Histologic Grading of Noninvasive Papillary Urothelial Neoplasms

european urology 51 (2007) 889 898 available at www.sciencedirect.com journal homepage: www.europeanurology.com Review Bladder Cancer Histologic Grading of Noninvasive Papillary Urothelial Neoplasms Gregory T. MacLennan a, Ziya Kirkali b, Liang Cheng c,d, * a Department of Pathology, Case Western Reserve University, Cleveland, OH, United States b Department of Urology, Dokuz Eylul University School of Medicine, Izmir, Turkey c Departments of Pathology and Laboratory Medicine, Indiana University School of Medicine, Indianapolis, IN, United States d Department of Urology, Indiana University School of Medicine, Indianapolis, IN, United States Article info Article history: Accepted October 17, 2006 Published online ahead of print on October 27, 2006 Keywords: Grading Neoplasia Papillary urothelial neoplasm of low malignant potential (PUNLMP) Transitional cell carcinoma Urinary bladder WHO/ISUP classification Abstract Objectives: In 1998, a revised system of classifying noninvasive papillary urothelial neoplasms of the urinary bladder was proposed and subsequently formally adopted by the World Health Organization (WHO). The introduction of this new system was justified as being potentially superior on a number of levels to the 1973 WHO classification system that it replaced. Specifically, a new category of neoplasms, designated papillary urothelial neoplasm of low malignant potential (PUNLMP), was considered advantageous for several reasons. The new system was expected to gain widespread acceptance, improve reproducibility of diagnoses among pathologists, and enhance the correlation between urine cytology and tumor histology. We examine the history of the changes in terminology for these lesions, the relative merits of PUNLMP terminology, the extent to which the expectations accompanying the new grading system have been met, and the extent to which the new system has enhanced the management of patients with noninvasive papillary urothelial neoplasms of the bladder. Methods: A PubMed literature search after the introduction of this new classification was performed and relevant papers reviewed. Results and Conclusions: The 2004 WHO classification is a positive initiative in attempting to standardize urothelial tumor grading by expanding and clearly defining the morphologic characteristics of noninvasive papillary urothelial neoplasms. The new terminology used in this system is of questionable validity and utility. Full-genome searches for prognostic and predictive molecular gene expression signatures, GeneChip technology and proteomics techniques, and several new biomarkers and molecular tests may be useful in future grading schemes after their clinical utility is better established. # 2006 European Association of Urology. Published by Elsevier B.V. All rights reserved. * Corresponding author. Departments of Pathology and Laboratory Medicine, Indiana University School of Medicine, 350 West 11th Street, Clarian Pathology Laboratory, Room 4010, Indianapolis, IN 46202, United States. Tel. +1 317 491 6442; Fax: +1 317 491 6419. E-mail address: lcheng@iupui.edu (L. Cheng). 0302-2838/$ see back matter # 2006 European Association of Urology. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.eururo.2006.10.037

890 european urology 51 (2007) 889 898 1. Introduction Nearly half of all bladder tumors are noninvasive (stage pta) papillary neoplasms of urothelial origin. These tumors have been intensively investigated for many decades, and a number of concepts regarding their biologic behavior and prognosis have been well established. Prognosis for these tumors is influenced by tumor size, tumor multifocality, recurrence status, coexistence of carcinoma in situ, and histologic tumor grade [1]. The first four elements are straightforward. However, there has been a longstanding lack of agreement among pathologists concerning the ideal system for grading these tumors. In this review, we examine this controversy, limiting our discussion to two grading systems, one of which (1973 World Health Organization [WHO]) was extensively used for several decades, and the other of which (2004 WHO) has been recommended as its replacement. 2. Historical perspective In the 1973 WHO classification of urothelial tumors, papillary urothelial neoplasms were separated into four categories: papilloma, and carcinoma grades 1 3 [2]. Papillomas were defined as exophytic tumors consisting of delicate fibrovascular cores covered by normal-looking urothelium with intact umbrella cells and virtually lacking mitotic activity. Clinically, these tumors are typically solitary and <1.0 cm in diameter, occur in patients <50 yr old, and have a negligible recurrence rate [3]. Regarding tumors diagnosed as carcinoma, histologic grading was based on the degree of cellular anaplasia, with grade 1 tumors having the least degree of anaplasia compatible with a diagnosis of malignancy, grade 3 tumors having the most severe degree of anaplasia, and grade 2 tumors having an intermediate degree of cellular anaplasia (Fig. 1) [2]. Anaplasia was further defined in the 1973 WHO classification as increased cellularity, nuclear crowding, disturbances of cellular polarity, failure of differentiation from the base to the surface, polymorphism, irregularity in the size of cells, variations of shape and chromatin pattern of the nuclei, displaced or abnormal mitotic figures, and giant cells [2]. A recurring criticism of the 1973 WHO grading system is that it did not define distinct cut-off points among the three tumor grades. At opposite ends of the spectrum of anaplasia, a pathologist may have little difficulty assigning a tumor as either grade 1 or grade 3 carcinoma. Distinguishing grade 2 carcinoma from grade 1 carcinoma at one end of the Fig. 1 Histologic grading of urothelial tumors. (A) Papillary urothelial neoplasm of low malignant potential (PUNLMP), formerly 1973 World Health Organization (WHO) grade 1 urothelial carcinoma; (B) low-grade urothelial carcinoma, formerly 1973 WHO grade 2 urothelial carcinoma; and (C) high-grade urothelial carcinoma, formerly 1973 WHO grade 3 urothelial carcinoma.

european urology 51 (2007) 889 898 891 Fig. 2 Comparison of the 1973 and 2004 WHO grading system. The 1973 WHO grade 1 carcinomas are reassigned, some to the PUNLMP category, and some to the low-grade carcinoma category. Similarly, 1973 WHO grade 2 carcinomas are reassigned, some to the low-grade carcinoma category, and others to the high-grade carcinoma category. All 1973 WHO tumors are assigned to the high-grade carcinoma category. WHO = World Health Organization; PUNLMP = papillary urothelial neoplasm of low malignant potential. spectrum and from grade 3 carcinoma at the opposite end of the spectrum (Fig. 2) is the most difficult aspect of applying the 1973 WHO grading system, with the result that there is wide variation in the reported frequency of grade 2 carcinoma, with reported incidences ranging from 13% to 69% [4]. This has raised concerns about lack of reproducibility in assigning tumor grades, and with this came concerns about the appropriateness of certain clinical management stratagems in a setting of uncertainty about proper tumor grade. Despite its apparent shortcomings, the 1973 WHO grading system has been in widespread use for more than three decades. It is accepted by uropathologists and uro-oncologists on a global scale. Enormous amounts of data have been accumulated using this system in studies of the morphologic properties, clinical behavior, treatment, and follow-up of urothelial tumors. The system has become well understood by clinicians, who are able to tailor patient management according to the reported grades. In the opinion of many, therefore, this grading system has never been broken, and consequently there is no apparent need to fix it [5]. Nevertheless, there existed a perceived need to develop a more universally acceptable classification system for bladder neoplasia that could be used effectively by pathologists, urologists, and oncologists. Consequently, following an initial meeting in 1997 among pathologists, urologists, and basic scientists in Washington, DC, at which it was agreed that an attempt would be made to create such a system, several members of the International Society of Urological Pathology (ISUP) in 1998 proposed a new grading system, subsequently known as the 1998 WHO/ISUP system. This system was developed by a select group of urologic pathologists with little input from urologists, medical oncologists, or radiation therapists, and studies designed to evaluate the clinical and prognostic relevance of this classification scheme were performed only after its publication. A revised version of this system (the 1999 WHO/ISUP system) met with limited acceptance. At a consensus conference in 2001, the majority opinion of the participants was that the 1973 WHO grading system should remain the international standard for the classification and grading of urothelial papillary neoplasms [4]. In 2004, a classification system for noninvasive papillary urothelial neoplasms, identical to the 1998 WHO/ISUP classification system, was adopted in Pathology and Genetics of Tumours of the Urinary System and Male Genital Organs, one of a series of WHO Blue Books for the classification of tumors. This new system separates noninvasive papillary urothelial neoplasms into four categories, designated papilloma, papillary urothelial neoplasm of low malignant potential (PUNLMP), low-grade carcinoma, and high-grade carcinoma. The recommendations in this book reflect the views of a Working Group of urologic pathologists assembled at an Editorial and Consensus Conference held in Lyon, France, in December 2002. The authors of the new 2004 WHO classification system expressed hopes and expectations that the new system would be widely accepted among all physicians affected by the system. A striking feature of the new system was the introduction of a newly designated category, PUNLMP, to circumvent use of the term carcinoma for tumors with a low probability of progression, but yet not entirely benign. It was emphasized in the introduction of this new system that it provides detailed histologic criteria for the diagnosis of papillary urothelial neoplasms, a feature that was expected to improve diagnostic reproducibility among pathologists. Additionally, it

892 european urology 51 (2007) 889 898 was anticipated that there would be reasonable consistency with the terminology used in urinary cytology, allowing easier cytohistologic correlation and improved patient management. In this review, we examine what has been learned about the usefulness of the 2004 WHO classification system for noninvasive papillary urothelial neoplasms since its introduction and attempt to ascertain whether it is indeed an improvement over the 1973 WHO system. Since the definition of papilloma is noncontroversial and is identical between the two systems, this entity will not be discussed. 3. Are the 1973 WHO and 2004 WHO classification systems interchangeable? The 2004 WHO classification system divides noninvasive papillary urothelial tumors into four categories as noted above. The definition for papilloma is the same in both the 1973 WHO and the 2004 WHO systems [6]. The authors of the new classification system have emphasized that the three newly introduced categories (PUNLMP, low-grade carcinoma, and high-grade carcinoma) are not directly interchangeable with the three 1973 WHO grades of carcinoma. The 1973 WHO grade 1 carcinomas are reassigned, some to the PUNLMP category, and some to the low-grade carcinoma category. Similarly, 1973 WHO grade 2 carcinomas are reassigned, some to the low-grade carcinoma category, and others to the high-grade carcinoma category. All 1973 WHO tumors are assigned to the high-grade carcinoma category. However, there is a fallacy in this reassignment process. On the one hand, it is implied that the descriptions of the original three WHO 1973 grades of carcinoma were so vague and subjective that they had virtually no useful definition; on the other hand, it is emphasized that the criteria for the three new categories are such that they do not allow direct correspondence with the original WHO 1973 grading criteria. If one set of entities (2004 WHO) has clear-cut definitions, and the other set of entities (1973 WHO) does not have clear-cut definitions, how then is it possible to state with certainty that the two different sets of entities are not essentially interchangeable? 4. What is a PUNLMP and what is its biologic behavior? This lesion is histologically defined by the 2004 WHO classification system as a papillary urothelial tumor that resembles the exophytic urothelial papilloma but shows increased cellular proliferation exceeding the thickness of normal urothelium (Fig. 1) [7]. It is characterized by papillary structures lined by an orderly arrangement of urothelial cells with minimal architectural abnormalities and minimal nuclear atypia [8]. Most, if not all, of these tumors would be diagnosed as grade 1 urothelial carcinomas using the 1973 WHO grading system. Cytologic atypia is minimal or absent and architectural abnormalities are minimal with preserved polarity. Mitotic figures are infrequent and are usually limited to the basal layer. Introduction of the PUNLMP category was designed to encompass a subset of bladder tumors of limited biologic aggressiveness, yet with features necessitating ongoing clinical follow-up. Another stated reason for its introduction was the intent to avoid burdening the patient with the psychosocial and financial (insurance) implications that come with a diagnosis of cancer. Shortly after the introduction of the term PUNLMP in the 1998 WHO/ISUP classification, Cheng reported a series of 112 patients whose bladder tumors showed findings consistent with PUNLMP, with up to 35 yr of follow-up (median, >12 yr), with tumor recurrence in 29% of patients [9]. Seventy-five percent of patients with tumor recurrence had a higher tumor grade, either low-grade or high-grade carcinoma according to the 1998 WHO/ISUP classification. The disease progression rate was 4% [9]. In other studies, tumor recurrence rates after a diagnosis of PUNLMP were reported to be35%inthestudybyholmangetal.[10] and 47% in the study by Pich et al. [11]. Samaratunga et al. found that PUNLMP had a progression rate of 8%, compared to a progression rate of 13% in low-grade carcinomas [12]. Fujii and associates reported a series of 53 patients with PUNLMP followed for a mean of 11.7 yr, with recurrences in 60%, grade migration to low-grade carcinoma in 34%, and progression to invasive carcinoma in 8% [13]. Both Holmang et al. and Oosterhuis et al. concluded that there was little if any significant difference in the risk of disease progression between PUNLMP and low-grade carcinoma [10,14,15]. Samaratunga and colleagues [12], in a study of 134 patients with noninvasive papillary urothelial tumors, observed that both the 1973 WHO and 2004 WHO grading systems were useful in predicting patient outcome ( p = 0.003 and p = 0.002, respectively). In their series, 8% of patients with PUNLMP experienced progression to invasive carcinoma, the highest PUNLMP progression rate of any published study [9,16]. During a median follow-up period of 56 mo,

european urology 51 (2007) 889 898 893 11% of patients with 1973 WHO grade 1 tumors experienced a progression rate of 11%, whereas 8% of those with PUNLMP tumors had progression. It is evident that tumors diagnosed as PUNLMP carry a substantial risk of tumor recurrence and disease progression. This is not surprising, considering that at least 80% of PUNLMP tumors can be demonstrated to carry allelic losses in putative tumor suppressor genes at multiple chromosomal loci that are comparable to those found in urothelial carcinoma [17]. Furthermore, point mutations in the FGFR3 gene were in 85% of PUNLMP tumors and in 88% of low-grade carcinomas in another study [18]. 5. Is PUNLMP actually carcinoma? Considering these studies of the biologic behavior and the molecular characteristics of PUNLMP, it seems evident that PUNLMP is an indolent variety of what we generally regard as carcinoma. A diagnosis of PUNLMP implies a real and significant potential for an adverse clinical outcome that does not differ greatly from that of 2004 WHO low-grade noninvasive urothelial carcinomas. By convention, appropriately or otherwise, the term carcinoma is routinely used to describe noninvasive neoplasms in the urinary bladder, a fact that generates confusion and controversy in the realm of proper histologic grading of bladder tumors. It is notable that the term carcinoma or adenocarcinoma is also routinely used to describe tumors without evidence of invasion in other organ systems, with and without the additional descriptor in situ. A small proportion of endometrial epithelial malignancies show no invasion into the underlying myometrium and have virtually no risk of dissemination, yet are characterized as endometrial adenocarcinoma. The authors of the 2004 WHO classification system are critical of the fact that very low-grade noninvasive tumors with a low probability of progressing are diagnosed in the 1973 WHO system as carcinomas, and this is regarded as being disadvantageous because it purportedly confers psychosocial stigmata and exposes patients to financial (insurance) hardships because they are labeled as having cancer. This criticism is irrelevant for citizens of countries that provide universal health coverage and for Americans with Medicare/Medicaid health coverage or group insurance. Even for Americans who have neither group insurance nor Medicare/Medicaid, it is questionable whether a diagnosis of PUNLMP is regarded by insurance carriers as any less ominous than a diagnosis of carcinoma. As detailed above, PUNLMPs have reported recurrence rates of up to 60% and progression rates of up to 8% [12 15,19]. This information is readily available to insurance carriers and their medical consultants, essentially invalidating the concept that a simple change in terminology is of any substantial financial benefit to patients. Speculations about the psychosocial burdens of a diagnosis of carcinoma as compared with a diagnosis of PUNLMP are unproven and possibly unfounded. In considering the above, it becomes difficult to accept the rationale behind retaining three categories of noninvasive papillary urothelial tumors in the new 2004 WHO system but giving them new and unfamiliar names, with the intent of calling two of the entities carcinoma, and being ambivalent about the third category (PUNLMP), which by many criteria also seems to qualify as a carcinoma. 6. Does the 2004 WHO classification system improve reproducibility? All grading systems are hampered by varying degrees of subjectivity that affect interobserver reproducibility. For example, Coblentz et al. found that 18% of bladder specimens with a referring diagnosis of urothelial carcinoma were found to have significant differences in diagnosis, stage, grade, or histologic tumor types when a second pathologist rendered an opinion [20]. Published reports of the reproducibility of grading systems are often derived from the efforts of small groups of pathologists who have previously worked or trained together; consequently, interobserver variation among pathologists unfamiliar with one another may be even greater than between such small groups. An important goal of the 2004 WHO classification was to provide detailed explanations of the histologic criteria for each diagnostic category, and thereby, to improve reproducibility among different pathologists. Despite provision of detailed histologic criteria for the diagnostic categories in the 2004 WHO system, improvement in intraobserver and interobserver variability as compared to the 1973 WHO system has not been documented. In fact, Mikuz et al. demonstrated that interobserver agreement was higher using the 1973 WHO classification than when using either the 2004 WHO or 1999 WHO/ISUP systems [21]. In a study by Yorokoglu and colleagues [22], the intraobserver and interobserver reproducibility of both the 2004 WHO and the 1973 WHO systems were evaluated by assigning six urologic pathologists to

894 european urology 51 (2007) 889 898 the task of independently reviewing 30 slides of noninvasive papillary urothelial tumors in a study set. They found no statistical difference between the reproducibility achieved with either system; the new system failed to improve reproducibility [22]. There was agreement for PUNLMP in only 48% of cases and reproducibility was lower for low-grade tumors in both the 2004 WHO and the 1973 WHO systems [22]. Murphy et al. recorded a 50% discrepancy rate among pathologists attempting to distinguish between PUNLMP and low-grade papillary urothelial carcinoma after a period of structured pathologist education [23]. In fairness, reproducibility in the 1973 WHO classification of urothelial tumors is also problematic. A frequent criticism of this classification scheme is that the morphologic criteria proposed for grading these neoplasms are vague and ill-defined, particularly those that are used in separating the three grades of carcinoma. No distinct cut-off points between the three tumor grades were defined in the 1973 WHO classification system, resulting in lack of agreement among pathologists concerning the proper assignment of grade in noninvasive urothelial tumors. At opposite ends of the spectrum of anaplasia, a pathologist may have little difficulty assigning a tumor as either grade 1 or grade 3 carcinoma. Distinguishing grade 2 carcinoma from grade 1 carcinoma at one end of the spectrum and from grade 3 carcinoma at the opposite end of the spectrum is the most difficult aspect of applying the 1973 WHO grading system, with the result that there is wide variation in the reported frequency of grade 2 carcinoma, with reported incidences ranging from 13% to 69%, and this is often mentioned as a major shortcoming of the WHO 1973 grading system, one that justifies the creation of a new improved grading system [4]. It is ironic, however, to note that in five separate publications in which noninvasive papillary urothelial tumors were graded according to the WHO/ISUP system, the incidence of PUNLMP varied from 12% to 39%, that of low-grade carcinoma varied from 27% to 63%, and the incidence of high-grade carcinoma varied from 21% to 67% [12,24 27]. Unfortunately, both grading systems are hampered by the fact that there is considerable heterogeneity within papillary urothelial neoplasms, making the assignment of a single grade to a tumor problematic in many instances [16]. In summary, hopes and expectations that the more precise histologic grading criteria specified in the new 2004 WHO classification would improve interobserver reproducibility do not appear to have been realized, based on analysis of data reported thus far. 7. Has the 2004 WHO grading system enhanced diagnostic accuracy in urine cytology? The authors of the 2004 WHO classification acknowledged that PUNLMP tumors would be difficult to diagnose by urinary cytopathology because of their minimal cytologic atypia. They postulated that the majority of 2004 WHO low-grade carcinomas could be diagnosed by this method [8]. Whisnant et al. correlated histologic findings in 86 transurethrally resected specimens representing a wide spectrum of papillary urothelial neoplasms with their corresponding urine cytology specimens [24]. They concluded that urine cytology in the context of the 1998 WHO/ISUP system (which is identical to the 2004 WHO system) is sensitive in detecting abnormal cells from PUNLMP and urothelial carcinoma but is not specific in distinguishing PUNLMP from low-grade carcinoma [24]. Curry et al. correlated the histologic findings with corresponding urine cytology findings in 100 randomly collected bladder tumor resections over a 3-yr period [27]. They concluded that, contrary to what had been predicted by the authors of the 1998 WHO/ ISUP system, the diagnostic sensitivity of urine cytology for the detection of low-grade urothelial neoplasms did not improve with the restructuring of the histologic classification system. Similar falsepositive rates were evident for the cytologic detection of low-grade urothelial carcinomas in the 1998 WHO/ISUP system and grade 2 urothelial carcinomas diagnosed according to the 1973 WHO system. In summary, routine urine cytology has been shown to be ineffective in differentiating PUNLMP from low-grade carcinoma and has a low sensitivity in detecting low-grade carcinoma, leading to the conclusion that the introduction of a new classification system for papillary noninvasive urothelial tumors has had essentially no beneficial impact in the field of urinary cytopathology, as originally anticipated. 8. Has the 2004 WHO classification system facilitated changes in clinical management of papillary urothelial neoplasms? In the past few decades it has been and still is well understood by most practicing urologists that noninvasive papillary urothelial tumors of all WHO 1973 grades require follow-up to detect recurrence or progression, despite the fact that grade 1 tumors are characteristically associated with an excellent prognosis. The length of clinical follow-up, the frequency of surveillance cystoscopy,

european urology 51 (2007) 889 898 895 and the adjunctive use of intravesical instillations of bacillus Calmette-Guérin (BCG) or a variety of chemotherapeutic agents is influenced by many factors, including histologic grade, tumor size, tumor multiplicity, depth of tumor invasion, recurrence history, and apparent grade migration with recurrence [1,16,28 37]. Currently, both in North America and Europe, there is no uniformity in the clinical management of patients with noninvasive papillary urothelial tumors. Patients with PUNLMP and noninvasive low-grade carcinoma are typically treated by transurethral resection of their tumors and are subsequently monitored for recurrence or progression by regular cystoscopy. Although lowgrade noninvasive carcinoma was found to have a statistically significant higher progression rate than PUNLMP in the study by Samaratunga et al. (8% for PUNLMP versus 13% for low-grade noninvasive urothelial carcinoma) [12], the reported high incidences of recurrence (up to 60% [13]) and progression (up to 8% [12,13]) for PUNLMP suggest that it is prudent to follow patients with a diagnosis of PUNLMP in an identical manner to patients with a diagnosis of low-grade noninvasive carcinoma. Indeed, investigators studying the recurrence and progression rate of PUNLMP have recommended long-term clinical follow-up for patients with these lesions [9,12 15]. To our knowledge, there has not been a published recommended surveillance protocol for PUNLMP tumors that differs significantly from the standard surveillance for low-grade noninvasive urothelial carcinomas. Nor, to our knowledge, have there been any published recommendations for following noninvasive urothelial tumors diagnosed according to the 2004 WHO grading system, nor for the use of intravesical therapy, that vary significantly from the traditions long established for following comparable lesions diagnosed according to the 1973 WHO grading system. In short, those who are charged with following these lesions appear to have gained minimal benefit from the new grading system in terms of surveillance or intravesical therapy protocols. 9. Conclusions The new 2004 WHO classification is a positive initiative in attempting to standardize urothelial tumor grading by expanding and clearly defining the morphologic characteristics of noninvasive papillary urothelial neoplasms. The new terminology used in this system is of questionable validity and utility. The creation of the category of papillary urothelial neoplasm of low malignant potential (PUNLMP) is especially contentious. It is unproven that this designation is any less a psychosocial stigma than a diagnosis of carcinoma. Assertions that this terminology is beneficial to patients because it helps them avoid financial/insurance complications of a diagnosis of carcinoma apply only to a relatively minute subset of patients on a global scale. The most disturbing aspect of this new term is that it has been demonstrated by multiple investigators to share biologic behavioral characteristics of carcinoma, and its molecular characteristics are identical to those of carcinoma, so it is difficult to understand why it is not being called carcinoma. The initial expectations that accompanied the creation of the new classification system have not been substantially met. Anticipation that the more precise histologic grading criteria specified in the new 2004 WHO classification would improve interobserver reproducibility appear to have been unfulfilled. Routine urine cytology has been shown to be ineffective in differentiating PUNLMP from lowgrade carcinoma, and has a low sensitivity in detecting low-grade carcinoma, leading to the conclusion that the introduction of a new classification system for papillary noninvasive urothelial tumors has had essentially no beneficial impact in the field of urinary cytopathology, as originally anticipated. Finally, in terms of clinical management of patients with noninvasive papillary urothelial neoplasms, there have been, to our knowledge, no published recommendations for following noninvasive urothelial tumors diagnosed according to the 2004 WHO grading system, nor for the use of intravesical therapy, that vary from the traditions long established for following comparable lesions diagnosed according to the 1973 WHO grading system. If the original focus in 1998 had been simply to retain the original 1973 WHO classification scheme, but to embellish it by expanding and clearly defining the morphologic characteristics of the original three grades of carcinoma, a long and unproductive period of controversy and uncertainty about the questionable merits of the new terminology could have been avoided. It is the opinion of the authors of this review that the 1973 WHO grading system should be re-established as the international standard for the classification and grading of urothelial papillary neoplasms. We do not favor adopting a two-grade system (low-grade versus high-grade) because we believe that, in combination with a consideration of other relevant factors, a three-grade system allows urologists more flexibility in devising treatment stratagems for individual patients. The three-grade system may also reflect

896 european urology 51 (2007) 889 898 better the inherent cancer heterogeneity of bladder tumors [16]. What does the future hold for grading bladder tumors? Investigations ranging from nuclear DNA analysis [38] to full-genome searches for prognostic and predictive molecular gene expression signatures as cancer markers have been performed in bladder tumors and have shown significant promise. Additionally, GeneChip technology and proteomics techniques may help to stratify patients into prognostic groups in the future. Several new biomarkers and molecular tests have shown promise in objectively differentiating patients at risk for a poor clinical outcome. It may be useful to eventually include these markers in future grading schemes after their clinical utility is better established. Conflicts of interest None of the authors (Gregory T. MacLennan, MD, Ziya Kirkali, MD, Liang Cheng, MD) have any commercial relationship such as consultancies, stock ownership or other equity interests, patents received and/or pending, or any commercial relationship which might be in any way considered related to a submitted article. References [1] Droller MJ. Bladder cancer: state-of-the-art care. CA Cancer J Clin 1998;48:269 84. [2] Mostofi FK, Sobin LH, Torloni H. Histological typing of urinary bladder tumours. Geneva, Switzerland: World Health Organization; 1973. [3] Cheng L, Darson M, Cheville JC, et al. Urothelial papilloma of the bladder: clinical and biologic implications. Cancer 1999;86:2098 101. [4] Bostwick DG, Mikuz G. Urothelial papillary (exophytic) neoplasms. Virchows Arch 2002;441:109 16. [5] Jones TD, Cheng L. Papillary urothelial neoplasm of low malignant potential: evolving terminology and concepts. J Urol 2006;175:1995 2003. [6] Lopez-Beltran A, Montironi R. Non-invasive urothelial neoplasms: according to the most recent WHO classification. Eur Urol 2004;46:170 6. [7] Eble JN, Sauter G, Epstein JI, Sesterhenn IAE, World Health Organization Classification of Tumours. Pathology and Genetics of Tumours of the Urinary System and Male Genital Organs. Lyon, France: IARC Press; 2004. p. 89 158. [8] Epstein JI, Amin MB, Reuter VR, Mostofi FK. The World Health Organization/International Society of Urological Pathology consensus classification of urothelial (transitional cell) neoplasms of the urinary bladder. Bladder Consensus Conference Committee. Am J Surg Pathol 1998;22:1435 48. [9] Cheng L, Neumann RM, Bostwick DG. Papillary urothelial neoplasms of low malignant potential: clinical and biological implications. Cancer 1999;86:2102 8. [10] Holmang S, Andius P, Hedelin H, Wester K, Busch C, Johansson SL. Stage progression in Ta papillary urothelial tumors: relationship to grade, immunohistochemical expression of tumor markers, mitotic frequency and DNA ploidy. J Urol 2001;165:1124 30. [11] Pich A, Chiusa L, Formiconi A, et al. Proliferative activity is the most significant predictor of recurrence in noninvasive papillary urothelial neoplasms of low malignant potential and grade 1 papillary carcinomas of the bladder. Cancer 2002;95:784 90. [12] Samaratunga H, Makarov DV, Epstein JI. Comparison of WHO/ISUP and WHO classification of noninvasive papillary urothelial neoplasms for risk of progression. Urology 2002;60:315 9. [13] Fujii Y, Kawakami S, Koga F, Nemoto T, Kihara K. Longterm outcome of bladder papillary urothelial neoplasms of low malignant potential. BJU Int 2003;92:559 62. [14] Holmang S, Hedelin H, Anderstrom C, Holmberg E, Busch C, Johansson SL. Recurrence and progression in low grade papillary urothelial tumors. J Urol 1999;162:702 7. [15] Oosterhuis JW, Schapers RF, Janssen-Heijnen ML, Pauwels RP, Newling DW, ten Kate F. Histological grading of papillary urothelial carcinoma of the bladder: prognostic value of the 1998 WHO/ISUP classification system and comparison with conventional grading systems. J Clin Pathol 2002;55:900 5. [16] Cheng L, Neumann RM, Nehra A, Spotts BE, Weaver AL, Bostwick DG. Cancer heterogeneity and its biologic implications in the grading of urothelial carcinoma. Cancer 2000;88:1663 70. [17] Cheng L, MacLennan GT, Zhang S, Wang M, Pan CX, Koch MO. Laser capture microdissection analysis reveals frequent allelic losses in papillary urothelial neoplasm of low malignant potential of the urinary bladder. Cancer 2004;101:183 8. [18] van Rhijn BW, Montironi R, Zwarthoff EC, Jobsis AC, van der Kwast TH. Frequent FGFR3 mutations in urothelial papilloma. J Pathol 2002;198:245 51. [19] Pich A, Chiusa L, Formiconi A, Galliano D, Bortolin P, Navone R. Biologic differences between noninvasive papillary urothelial neoplasms of low malignant potential and low-grade (grade 1) papillary carcinomas of the bladder. Am J Surg Pathol 2001;25:1528 33. [20] Coblentz TR, Mills SE, Theodorescu D. Impact of second opinion pathology in the definitive management of patients with bladder carcinoma. Cancer 2001;91: 1284 90. [21] Mikuz G. The reliability and reproducibility of the different classifications of bladder cancer. In: Haupmann S, Dietel M, Sorbrinho-Simoes M, editors. Surgical pathology update 2001. Berlin: ABW-Wissenschaftsverlag; 2001. p. 114 5. [22] Yorukoglu K, Tuna B, Dikicioglu E, et al. Reproducibility of the 1998 World Health Organization/International Society

european urology 51 (2007) 889 898 897 of Urologic Pathology classification of papillary urothelial neoplasms of the urinary bladder. Virchows Arch 2003;443:734 40. [23] Murphy WM, Takezawa K, Maruniak NA. Interobserver discrepancy using the 1998 World Health Organization/ International Society of Urologic Pathology classification of urothelial neoplasms: practical choices for patient care. J Urol 2002;168:968 72. [24] Whisnant RE, Bastacky SI, Ohori NP. Cytologic diagnosis of low-grade papillary urothelial neoplasms (low malignant potential and low-grade carcinoma) in the context of the 1998 WHO/ISUP classification. Diagn Cytopathol 2003;28:186 90. [25] Bircan S, Candir O, Serel TA. Comparison of WHO 1973, WHO/ISUP 1998, WHO 1999 grade and combined scoring systems in evaluation of bladder carcinoma. Urol Int 2004;73:201 8. [26] Yin H, Leong AS. Histologic grading of noninvasive papillary urothelial tumors: validation of the 1998 WHO/ISUP system by immunophenotyping and follow-up. Am J Clin Pathol 2004;121:679 87. [27] Curry JL, Wojcik EM. The effects of the current World Health Organization/International Society of Urologic Pathologists bladder neoplasm classification system on urine cytology results. Cancer 2002;96:140 5. [28] Cheng L, Neumann RM, Weaver AL, et al. Grading and staging of bladder carcinoma in transurethral resection specimens. Correlation with 105 matched cystectomy specimens. Am J Clin Pathol 2000;113:275 9. [29] Cheng L, Bostwick DG. World Health Organization and International Society of Urological Pathology classification and two-number grading system of bladder tumors: reply. Cancer 2000;88:1513 6. [30] Cheng L, Neumann RM, Weaver AL, Spotts BE, Bostwick DG. Predicting cancer progression in patients with stage T1 bladder carcinoma. J Clin Oncol 1999;17:3182 7. [31] Cheng L, Neumann RM, Scherer BG, et al. Tumor size predicts the survival of patients with pathologic stage T2 bladder carcinoma: a critical evaluation of the depth of muscle invasion. Cancer 1999;85:2638 47. [32] Cheng L, Weaver AL, Leibovich BC, et al. Predicting the survival of bladder carcinoma patients treated with radical cystectomy. Cancer 2000;88:2326 32. [33] Cheng L, Cheville JC, Leibovich BC, et al. Survival of patients with carcinoma in situ of the urinary bladder. Cancer 1999;85:2469 74. [34] Cheng L, Cheville JC, Neumann RM, Bostwick DG. Natural history of urothelial dysplasia of the bladder. Am J Surg Pathol 1999;23:443 7. [35] Cheng L, Cheville JC, Neumann RM, Bostwick DG. Flat intraepithelial lesions of the urinary bladder. Cancer 2000;88:625 31. [36] Cheng L, Weaver AL, Bostwick DG. Predicting extravesical extension of bladder carcinoma: a novel method based on micrometer measurement of the depth of invasion in transurethral resection specimens. Urology 2000;55: 668 72. [37] Cheng L, Weaver AL, Neumann RM, Scherer BG, Bostwick DG. Substaging of T1 bladder carcinoma based on the depth of invasion as measured by micrometer. A new proposal. Cancer 1999;86:1035 43. [38] Montironi R, Scarpelli M, De Nictolis M, Mariuzzi G, Ansuini G, Pisani E. Comparison of computerized analysis of nuclear DNA changes in uterine cervix dysplasia and in urothelial non-invasive papillary carcinoma. Pathol Res Pract 1988;183:489 96. Editorial Comment Rodolfo Montironi, Polytechnic University of the Marche Region, Ancona, Italy r.montironi@univpm.it The classification of the noninvasive urothelial papillary lesions used worldwide is the 1973 World Health Organization (WHO) classification. Urologists have learned what to do in terms of therapy and follow-up. The classification has been criticised for several reasons, including that what is called grade 1 (G1) papillary carcinoma is not really a carcinoma. To me the problem of G1 not being a carcinoma is related to the fact that the noninvasive urothelial lesions other than papilloma are classified as papillary carcinomas (grades 1, 2, and 3), whereas lesions with similar morphology but without a papillary configuration (i.e., flat) are called hyperplasia, dysplasia, and carcinoma in situ. Should we use an identical approach for papillary and flat lesions? What is called G1 papillary carcinoma has the features of a lesion that could be called papillary hyperplasia and papillary dysplasia. The latter includes probably also what was in the past defined as grade 2A papillary carcinoma. The remaining part of the spectrum has the features of papillary carcinoma in situ [1]. The so-called 2004 WHO classification (formerly International Society of Urological Pathology [IISUP]/WHO 1998) was supposed to replace the 1973 classification. Urologists have been faced with at least two problems. One is the difficulty in understanding what papillary urothelial neoplasm of low malignant potential (PUNLMP) is. I have seldom heard that, to some urologists, PUNLMP means something like atypical small acinar proliferation (ASAP) of the prostate, which it is not. The other problem is that because there is no complete agreement in the three grades of papillary carcinoma of the 1973 classification and the three lesions of the 2004 classification, urologists

898 european urology 51 (2007) 889 898 might not exactly know what they have to do in patients with PUNLMP or with low-grade or highgrade papillary carcinoma. Pathologists have their problems when the new classification is applied. Some recent studies have shown that diagnostic reproducibility is disturbingly low [2]. The proposal made in the 2001 Ancona Consultation remains valid [3]: both classifications should be used in the pathology reports and let urologists select the one they understand better or feel more confident using. References [1] Montironi R, Scarpelli M, De Nictolis M, Mariuzzi G, Ansuini G, Pisani E. Comparison of computerized analysis of nuclear DNA changes in uterine cervix dysplasia and in urothelial non-invasive papillary carcinoma. Pathol Res Pract 1988;183:489 96. [2] Park YW, Shim SI, Lee DW, et al. ISUP/WHO classification of urothelial neoplasms of urinary bladder. Mod Pathol 2006;9(Suppl 3):86A. [3] Bostwick DG, Mikuz G. Urothelial papillary (exophytic) neoplasms. Virchows Arch 2002;441:109 16.