Jeremy S. Miller

Non-invasive papillary urothelial neoplasms: The 2004 WHO/ISUP classification system

The classification and grading of papillary urothelial neoplasms has been a long‐standing subject of controversy. Previously, numerous diverse grading schemes for bladder tumor, including the 1973 World Health Organization (WHO) classification, existed whereby one of the major limitations was poor inter‐observer reproducibility among pathologists. The WHO/International Society of Urological Pathology (ISUP) consensus classification system of urothelial neoplasms of the urinary bladder was developed in 1998 and was revised most recently in 2003 (published in 2004). Importantly, the current classification system provides detailed histological criteria for papillary urothelial lesions and allows for designation of a lesion (papillary urothelial neoplasm of low malignant potential) with a negligible risk of progression. Thus, the latest system is designed to be a universally acceptable one for bladder tumors that not only could be effectively used by pathologists, urologists, and oncologists, but also stratifies the tumors into prognostically significant categories. This article outlines the 2004 WHO/ISUP classification system regarding the specific histological criteria for non‐invasive papillary urothelial neoplasms and the clinical significance of each category.

Small Endoscopic Biopsies of the Ureter and Renal Pelvis: Pathologic Pitfalls

Technical advances in endoscopic equipment have led to increased ureteroscopic biopsies of the upper urinary tract, resulting in limited biopsy material. We retrospectively reviewed 76 consecutive mid-upper ureter and renal pelvis biopsies submitted for consultation from January 2004 to January 2009, where follow-up was obtainable. There were 49 (64.5%) males and 27 (35.5%) females. Thirty-nine (51.3%) of the biopsies were from the ureter with the remaining 37 (48.7%) from the renal pelvis. The mean age was 70 years for males and 71 for females (range: 24 to 89). At consultation, the most common diagnoses were benign urothelium (n=25, 32.9%); atypical (n=17, 22.4%); low-grade noninvasive papillary urothelial carcinoma (n=10, 13.2%); and high-grade noninvasive papillary urothelial carcinoma (n=8, 10.5%). In cases where a definitive diagnosis could not be reached on expert review, it was mainly because of the limited size of the biopsy, absence of papillary fronds, crush artifact, and distorted architecture. There were 7 major discrepancies between the outside and second opinion diagnosis, where all of the cases were initially diagnosed as an urothelial neoplasm, yet was non-neoplastic upon review. Strips of urothelium without well-developed fibrovascular cores, polypoid ureteritis/pyelitis, and reactive urothelium mimicked urothelial neoplasms. In 5 of these 7 cases, there was no gross lesion suspicious of a tumor present according to the urologist. Overall, 33 of the 44 (75%) cases with a mass noted by the urologist or by radiography was found to have a neoplasm at follow-up. Conversely, 24 of the 32 (75%) cases without a grossly suspected tumor had no neoplasm at follow-up. The association between the histologic presence of a neoplasm at follow-up and the presence of a clinically suspected tumor was highly significant (P<0.0001). Pathologists need to recognize that in almost 1 of the 4 renal pelvic/ureteral biopsies a definitive diagnosis cannot be made because of the inadequate tissue. Caution must be exercised in the evaluation of these limited specimens, especially in the absence of a clinically suspected tumor.

Low-grade papillary urothelial carcinoma of the urinary bladder: a clinicopathologic analysis of a post-World Health Organization/International Society of Urological Pathology classification cohort from a single academic center.

CONTEXT Few large cohort studies have addressed outcome in patients with noninvasive low-grade papillary urothelial carcinoma (LG-UrCa) following implementation of the 2004 World Health Organization/International Society of Urological Pathology (WHO/ISUP) consensus classification. OBJECTIVE To evaluate our cohort of LG-UrCa cases classified according to 2004 WHO/ISUP to reassess outcome and interobserver agreement. DESIGN Files were searched for all patients diagnosed with LG-UrCa between 1998 and 2008. All sections were reevaluated for accuracy of classification. RESULTS A total of 112 cases initially diagnosed as LG-UrCa were identified. Of those, 8 of 55 cases (15%) initially diagnosed by nonurologic pathologists were reclassified as high-grade papillary urothelial carcinoma and were excluded. The mean length of follow-up was 40.1 months (range, 2-113 months). Tumor recurrence was encountered in 56 of 104 patients (53.8%), including 37 (35.6%) with LG-UrCa or lower-grade tumors and 19 (18.3%) with high-grade papillary urothelial carcinoma. Of the 19 patients demonstrating grade progression, 7 (37%) also developed stage progression (invasive carcinoma, n = 5; metastatic carcinoma, n = 2). Seven patients eventually underwent radical cystectomy. None of the 104 patients died of bladder cancer. The mean number of recurrence episodes was 3.11. The mean durations of time to first recurrence and time to grade progression were 13.9 months and 25.1 months, respectively. The mean size of initial tumors was 1.73 cm. There was no significant correlation between tumor size, patient age, sex, or smoking history and the likelihood for recurrence or grade progression. A significantly higher rate of recurrence was seen in patients with multiple tumors at initial diagnosis (P = .04). CONCLUSIONS A tendency to underdiagnose high-grade papillary urothelial carcinoma continues to exist. More than half (53.8%) of patients with LG-UrCa developed recurrence, with an 18.3% incidence of grade progression and a 6.7% incidence of stage progression. Patients with multiple initial tumors had significantly higher risk of developing recurrence.

Primary leiomyosarcoma of the kidney: A clinicopathologic study of 27 cases

Primary leiomyosarcoma of the kidney is a rare entity that has not been well characterized. We retrieved 27 cases of primary renal leiomyosarcomas diagnosed at 3 institutions between 1986 and 2009. Mean patient age at diagnosis was 58.5 years (range 22 to 85), and 59% were female. Mean tumor size was 13.4 cm (range 4 to 26), and 59% of the tumors were identified in the right kidney. Detailed histologic examination was possible for 24 of the cases. Average mitotic count per 10 high-power fields was 11.1 (range 0 to 50), and the average extent of necrosis was 21% (range 0 to 60). Cellular pleomorphism was classified as either focal (n=13) or extensive (n=11) and graded as mild (n=3), moderate (n=7) or severe (n=14). Tumors were either grade 2 (n=12) or grade 3 (n=12) using the French Federation of Cancer Centers System. Direct extension beyond the kidney capsule was identified in 55% of the cases, and lymphovascular invasion was identified in 26%. Clinical follow-up information was available for 20 of the cases, and patients were followed for an average of 2.8 years (range 0.25 to 9). Distant metastases were identified in 90% of the patients, and 75% eventually died from their tumors burden. In conclusion, primary renal leiomyosarcomas have a grim prognosis regardless of the underlying histology.

Identification of gleason pattern 5 on prostatic needle core biopsy: Frequency of underdiagnosis and relation to morphology

The presence of a Gleason pattern 5 prostatic adenocarcinoma is associated with a worse outcome. This study assesses the accuracy of grading a tumor as having Gleason pattern 5 and the potential factors contributing to its undergrading. From the consultation service of one of the authors, we identified 59 consecutive needle biopsy cases comprising 138 parts that, upon review, were graded as having Gleason pattern 5. All cases were reported as the final diagnosis by the outside pathologist. They were sent for a second opinion at the behest of clinicians or patients and not because the pathologist was seeking a second opinion. Considering the highest Gleason score in a given multicore specimen as the overall Gleason score, Gleason pattern 5 was missed in 34 of 59 (57.6%) cases by the outside pathologist. Compared with the outside pathologist’s diagnosis, the Gleason score rendered at the second opinion was increased in 101 of 138 (73.2%) parts, was decreased in 5 of 138 (3.6%) parts, and remained unchanged in 32 of 138 (23.2%) parts. Gleason pattern 5 was not identified by the initiating pathologist in 67 of 138 (48.6%) of the evaluated parts. The architectural patterns of pattern 5 were as follows: single cells (n=104, 75.3%); solid sheets (n=69, 50%); cords (n=62, 44.9%); and comedonecrosis (n=3, 2.2%). Pattern 5 was missed more frequently when it was not the primary pattern. The most common Gleason pattern 5 architectural type was single cells and the least common was comedonecrosis. None of the architectural patterns appeared to be more correctly identified than the others; however, the most accurate grading was when the primary pattern was 5 and was composed mostly of solid sheets. Owing to the important prognostic and therapeutic implications of Gleason pattern 5, pathologists must be attuned to its varied patterns and to the fact that it may often represent a secondary or tertiary component of the carcinoma.

Noninvasive Urothelial Carcinoma of the Bladder With Glandular Differentiation: Report of 24 Cases

Noninvasive urothelial carcinoma (UC) with glandular differentiation in the absence of infiltrating carcinoma is a rare entity that has not been well characterized. We retrieved 24 cases of noninvasive UC of the bladder with glandular differentiation on biopsy (n=20) or transurethral resection (n=4) without an associated invasive component. The cases were identified from the consult files of one of the authors between 1992 and 2008. Mean patient age at diagnosis was 70 years (range: 48 to 87 y) and 75% were male. Half of the cases were pure noninvasive UC with glandular differentiation; half were associated with either carcinoma in situ or high-grade noninvasive papillary carcinoma. The glandular component consisted of 1 or more patterns: papillary (46% of cases), glandular (42%), cribriform (33%), and flat (25%). Mitoses, apoptosis, and necrosis were identified in 83%, 67%, and 17% of the biopsies, respectively. One case was a recent diagnosis, and 5 patients either refused treatment or were lost to follow-up. Of the 18 patients with available follow-up information, 9 (50%) did not develop invasive carcinoma; the remaining 9 (50%) eventually developed an invasive bladder tumor. Of these, 2 were small cell carcinoma, 3 were poorly-differentiated UC (2 of these developed widespread metastases), and 4 were UC, not otherwise specified. In both instances of eventual small cell carcinoma, and in 2 of the 3 cases of poorly-differentiated UC, the initial biopsy consisted of pure noninvasive UC with glandular differentiation without carcinoma in situ or noninvasive papillary carcinoma. Of note, none of the patients in the study developed invasive adenocarcinoma.

Rare histological patterns of prostatic ductal adenocarcinoma

Aims: Prostatic ductal adenocarcinomas account for 1% of prostate cancers. Most commonly, these lesions grow in large cribriform and/or papillary patterns or, as recently described, in a manner resembling prostatic intraepithelial neoplasia (i.e., ‘PIN‐like’ prostatic ductal adenocarcinoma). This study aims to report rare variants of ductal adenocarcinoma. Methods: Ten cases of rare patterns of prostatic ductal adenocarcinoma that have not been formally investigated prior to this study, primarily from one authors consultation service (1987–2009), were selected. Results: Two (n = 2) cases were prostatic ductal adenocarcinoma with mucinous and goblet cell features. Three (n = 3) cases are the first described cases of foamy gland prostatic ductal adenocarcinoma. Other unique cases were prostatic duct adenocarcinomas with associated Paneth cell‐like neuroendocrine (n = 2), micropapillary (n = 2), and cystic papillary features (n = 1). Prostatic origin was confirmed with immunohistochemical studies for prostate specific antigen (PSA), P501S, and prostate specific membrane antigen (PSMA). High‐grade PIN was ruled out with negative stains for high molecular weight cytokeratin (HMWCK) and p63. Four prostatic ductal adenocarcinomas had no evidence of disease at 2–8 years follow‐up: foamy gland, Paneth cell‐like, and micropapillary (two cases). One mucinous prostatic ductal adenocarcinoma resulted in the patients death and the other mucinous case was alive at 7 years and 2 months, yet with no information as to status of disease. The remaining four cases were lost to follow‐up, died of other causes, or were recent. Conclusions: In summary, we report several rare and unique histological patterns of prostatic ductal adenocarcinoma. The practical importance of recognising these histological variations is that in some cases they may be misdiagnosed as non‐prostatic tumours. These unusual cases also provide further support for the relationship between acinar and ductal adenocarcinoma.

Extraprostatic extension of prostatic adenocarcinoma on needle core biopsy: report of 72 cases with clinical follow-up

Study Type – Prognosis (case series)
Level of Evidence 4

Papillary urothelial neoplasm of low malignant potential of the urinary bladder: clinicopathologic and outcome analysis from a single academic center

Few long-term single-center studies have addressed the outcome of patients with papillary urothelial neoplasms of low malignant potential. Our study evaluates the behavior of these tumors occurring as primary urinary bladder lesions. For this purpose, 34 primary in-house cases diagnosed and treated between 1998 and 2008 were identified from our medical records. Upon review, 3 cases were upgraded to noninvasive low-grade urothelial carcinomas and excluded from further evaluation. During follow-up (range, 3-108 months; mean, 42 months), 13 patients developed recurrences; and 9 patients progressed to a noninvasive higher grade lesion (8 to low-grade and 1 to high-grade urothelial carcinomas). None of our patients developed stage progression (>pTa) or died of bladder cancer. Size of the primary tumor was associated with the risk of recurrence (P = .043), whereas the number of episodes of recurrence was associated with the likelihood of grade progression (P = .034). In conclusion, recurrences were observed in 42% of all our patients, with a grade progression rate of 29%. None of our patients developed invasive carcinoma or died as a consequence of their disease. Considering the low but definitive risk of recurrence and grade progression, appropriate clinical follow-up of patients with primary papillary urothelial neoplasm of low malignant potential is warranted.

The utility of microscopic findings and immunohistochemistry in the classification of necrotic testicular tumors: a study of 11 cases.

Necrotic testicular tumors are relatively frequent and can present a significant diagnostic challenge. Because of differing treatments for seminomas versus nonseminomas, accurate diagnosis is critical. Eleven totally (n=9) or almost totally (n=2) necrotic testicular tumors were retrieved from our consult files. The submitting pathologists favored benign processes in 4 cases, Leydig cell tumor in 1, and lymphoma in 1. The cases were evaluated for histologic features and, when material was available, by immunostaining with 7 antibodies: keratin (AE1/AE3), OCT4, placental alkaline phosphatase, α-fetoprotein (AFP), CD117, CD30, and S100. Only distinct reactivity in a cellular distribution in the necrotic zone was considered positive; nuclear reactivity alone was scored for OCT4 and membrane reactivity for CD117 and CD30. Mean patient age was 35 years (range 16-63). Mean tumor size was 19 mm (range 7-53). All patients presented with unilateral testicular masses (6 right, 5 left); 2 also had acute pain. The combination of histologic features, immunostains and, in 1 case, serum AFP permitted classification of 8 tumors (4 seminomas, 3 embryonal carcinomas, 1 yolk sac tumor). Three were not classifiable. The necrotic seminomas lacked associated coarse intratubular calcifications and were positive for OCT4 (4/4) and CD117 (3/3) but negative for keratin (0/4) and CD30 (0/4). The necrotic embryonal carcinomas had associated coarse intratubular calcifications and were positive for keratin (2/3), OCT4 (2/2), and CD30 (3/3). OCT4 stained 1 unclassifiable tumor, which lacked other specific markers. We did not find placental alkaline phosphatase, AFP, and S100 stains useful, although S100 did highlight tumor “ghost” cells in 1 case. Other features in most cases included intratubular germ cell neoplasia (6/11), tubular atrophy/hyalinization (10/11), tumor “ghost” cells (10/11), scar (9/11), and inflammation (10/11). Of the 5 patients with available follow-up, 3 were free of disease at 1, 5, and 8 years after orchiectomy (2 necrotic seminomas and 1 germ cell tumor, unclassified). One patient with yolk sac tumor (age 63 y) developed widespread metastases after 15 months and died of disease. The final case was initially misinterpreted as “testicular infarction, no malignancy” and 16 months later the patient developed a large retroperitoneal seminoma. Most totally necrotic testicular tumors can be placed into clinically important groups by assessment for coarse intratubular calcifications and staining reactions for keratin, OCT4, CD117, and CD30.