Laser capture microdissection analysis reveals frequent allelic losses in papillary urothelial neoplasm of low malignant potential of the urinary bladder

Authors

  • Liang Cheng M.D.,

    Corresponding author
    1. Department of Pathology and Laboratory Medicine, Indiana University School of Medicine, Indianapolis, Indiana
    2. Department of Urology, Indiana University School of Medicine, Indianapolis, Indiana
    • Department of Pathology and Laboratory Medicine, Indiana University Medical Center, University Hospital 3465, 550 North University Boulevard, Indianapolis, IN 46202
    Search for more papers by this author
    • Fax: (317) 274-5346

  • Gregory T. MacLennan M.D.,

    1. Department of Pathology, Case Western Reserve University, Cleveland, Ohio
    Search for more papers by this author
  • Shaobo Zhang M.D.,

    1. Department of Pathology and Laboratory Medicine, Indiana University School of Medicine, Indianapolis, Indiana
    Search for more papers by this author
  • Mingsheng Wang M.D.,

    1. Department of Pathology and Laboratory Medicine, Indiana University School of Medicine, Indianapolis, Indiana
    Search for more papers by this author
  • Chong-Xian Pan M.D., Ph.D.,

    1. Department of Medicine, Indiana University School of Medicine, Indianapolis, Indiana
    Search for more papers by this author
  • Michael O. Koch M.D.

    1. Department of Urology, Indiana University School of Medicine, Indianapolis, Indiana
    Search for more papers by this author

Abstract

BACKGROUND

In the 1999 World Health Organization classification system, papillary tumors of the urinary bladder were classified as papilloma, papillary urothelial neoplasm of low malignant potential (PUNLMP), and as Grade 1, Grade 2, and Grade 3 urothelial carcinoma. The biologic potential of PUNLMP of the urinary bladder is controversial. To the authors' knowledge, information regarding the genetic changes of PUNLMP tumors of the bladder is limited.

METHODS

The authors examined loss of heterogygosity (LOH) at 5 polymorphic microsatellite markers on chromosome 9q32–33 (D9S177), 9p22 (IFNA), 17p13.1 (TP53), 12q14–24 (D12S1051), and 3p25–26 (D3S3050) from 26 patients who were diagnosed with PUNLMP tumors of the urinary bladder. Tumors were microdissected from sections prepared from formalin-fixed, paraffin-processed tissue specimens using laser capture microdissection.

RESULTS

LOH was found in 21 of 26 (81%) patients with PUNLMP. The rate of LOH was 41% with D9S177, 32% with IFNA, 29% with TP53, 26% with D12S1051, and 44% with D3S3050. Allelic loss of multiple chromosome loci was often present in patients with PUNLMP tumors.

CONCLUSIONS

Genetic changes that commonly occur in advanced bladder carcinoma (≥ pT2) are frequently found in PUNLMP of the urinary bladder. Cancer 2004. © 2004 American Cancer Society.

The term “papillary urothelial neoplasm of low malignant potential” (PUNLMP) was first introduced in the 1998 World Health Organization/International Society of Urologic Pathology (WHO/ISUP) classification of bladder tumors,1 which was subsequently adopted in the 1999 and 2004 WHO classification of the urinary bladder.2 In the 1999 WHO classification system, papillary tumors were classified as papilloma, PUNLMP, and as Grade 1, Grade 2, and Grade 3 urothelial carcinoma,2 whereas the 1998 WHO/IUSP classification divided papillary bladder tumors into papilloma, PUNLMP, and low-grade and high-grade carcinoma. Unlike the previous WHO classifications,2, 3 the 2004 WHO classification of bladder carcinoma does not include Grade 1 urothelial carcinoma.4 Three distinct categories were used, namely, PUNLMP, low-grade, and high-grade urothelial carcinoma. The authors of the new classification acknowledge that it reflects work in progress and that the classification may undergo further refinements as new molecular, genetic, and clinical data become available.4

Patients with a diagnosis of urothelial papilloma have a low incidence of disease recurrence (8%) and the development of urothelial carcinoma is even more uncommon.5 In a large series of 52 patients with urothelial papilloma, only 1 patient developed Grade 1 noninvasive urothelial carcinoma during a mean follow-up of 9.8 years, none developed urothelial dysplasia, carcinoma in situ, or invasive urothelial carcinoma, and none died of bladder carcinoma.5 In contrast, a significant number of patients with PUNLMP have increased risk of developing local disease recurrence, progression, and cancer death.6, 7

Previous studies8–20 have shown that urothelial carcinoma is characterized by frequent loss of heterozygosity (LOH) in several different chromosomes, whereas LOH is uncommon in urothelial papilloma. Chow et al.21 studied four patients with urothelial papilloma and found that none had LOH from nine different chromosome regions. To our knowledge, little is known regarding the genetic alterations in PUNLMP. We examined LOH status in 26 patients with PUNLMP using laser capture microdissection.

MATERIALS AND METHODS

We studied 26 patients who underwent transurethral resection for bladder tumors. These patients were selected randomly from the surgical pathology files at the Indiana University Hospital (Indianapolis, IN) and University Hospitals of Cleveland (Cleveland, OH) between 1995 and 2002. Patients with insufficient tissue specimens for genetic analysis were excluded. Grading of the primary tumor was performed according to the 1999 WHO classification system for bladder carcinoma.1, 2 All slides were retrospectively reviewed and fulfilled the diagnostic criteria of PUNLMP.1, 2 None of the patients had coexisting urothelial dysplasia, carcinoma in situ, or invasive urothelial carcinoma. Patients had a mean age of 65 years (range, 33–88 years) and the male-to-female ratio was 4:1. The current study was approved by the institutional review boards at both hospitals.

Genomic DNA was prepared from primary tumor tissue specimens and adjacent normal tissue specimens using the PixCell II laser capture microdissection system (Arcturus, Mountain View, CA). Five polymorphic microsatellite markers on chromosome 9q32–33 (D9S177), 9p22 (IFNA), 17p13.1 (TP53), 12q14–24 (D12S1051), and 3p25–26 (D3S3050) were used. Allelic loss of these chromosome regions has been documented in patients with advanced bladder carcinoma.8–20 Polymerase chain reaction (PCR) amplification and gel electrophoresis were performed as previously described.22–27 LOH was determined by a combination of direct visualization and scanning densitometry of the autoradiograms with a phosphorimager (Molecular Dynamics Inc., Sunnyvale, CA). LOH was defined as complete or near complete absence of one allele (≥ 75% loss) in the tumor DNA.28 PCR techniques for each polymorphic microsatellite marker were repeated at least twice, and the same results were obtained.

RESULTS

Table 1 summarizes the LOH findings on chromosome regions frequently involved in advanced bladder carcinoma. Of 26 patients with PUNLMP, the pattern was informative for 22 patients at D9S177 locus, 25 patients at IFNA locus, 24 patients at TP53 locus, 23 patients at D12S1051 locus, and 18 patients at D3S3050 locus. The overall frequency of allelic loss was 81% in PUNLMP. The rate of LOH was 41% with D9S177, 32% with IFNA, 29% with TP53, 26% with D12S1051, and 44% with D3S3050. Ten patients demonstrated LOH at 1 locus, 7 patients demonstrated LOH at 2 loci, 2 patients demonstrated LOH at 3 loci, and 2 patients demonstrated LOH at 4 loci. None had homozygous deletion. Figure 1 illustrates the microdissection of PUNLMP and representative results of LOH analysis.

Table 1. Analysis of LOH in Patients with Papillary Urothelial Neoplasm of Low Malignant Potential
Patient no.StageLOH
D9S177IFNATP53D12S1051D3S3050
  1. LOH: loss of heterozygosity; NL: no allelic loss; NI: noninformative.

1TaLOHLOHNLNINL
2TaLOHNLNLNINL
3TaLOHLOHNLNLNI
4TaLOHLOHNLNLNL
5TaNINLNLNLNL
6TaLOHLOHLOHNLLOH
7TaNLNLNLNLLOH
8TaLOHLOHNLLOHNI
9TaNININLNLNL
10TaNINLLOHNILOH
11TaNLNLLOHNLNI
12TaNLNLNLNLLOH
13TaNLNLNINLLOH
14TaNLNLNLLOHNI
15TaNLNLNLNLNI
16TaNLNLLOHNLLOH
17TaLOHNLNLNLLOH
18TaNLNLNLLOHNL
19TaNLNLNLNLNI
20TaNINLNLNLNI
21TaNLLOHNLNLNL
22TaNLNLNLLOHNI
23TaLOHNLNILOHNL
24TaNLNLLOHNLNL
25TaLOHLOHLOHNLNL
26TaNLLOHLOHLOHLOH
Figure 1.

Microdissection of papillary urothelial neoplasm of low malignant potential and representative results of loss of heterozygosity (LOH) analysis. (A) Tumor specimen before microdissection. (B) Tumor specimen after microdissection. (C) Laser-captured tumor cells. (D) LOH results from the same patient. N: normal; T: tumor; arrow indicates allelic loss.

DISCUSSION

Carcinoma of the urinary bladder is a significant cause of cancer morbidity and mortality in the U.S. In 2004, it is estimated that 60,240 new cases and 444,640 deaths will occur.29 Surveillance and screening programs have increased the likelihood of detection of early-stage disease. A reliable and reproducible grading system is needed to stratify patients into prognostically distinct groups and to allow for comparison of treatment results from different institutions.

The 1973 WHO classification and grading system for urinary bladder tumors is accepted widely by pathologists, urologists, oncologists, and cancer registrars in the U.S. It classified papillary bladder tumors into four categories: papilloma, and as Grade 1, Grade 2, and Grade 3 urothelial carcinoma.3 Since its publication in 1973, the WHO grading system has been repeatedly validated around the world. Data accumulated in the world literature over the past 30 years indicate that the 1973 WHO grading system is one of most powerful predictors of patient outcome. The prognosis of patients with bladder carcinoma is directly linked to tumor grade.30–37 In the U.S., grading of urothelial carcinoma by pathologists is the most consistent of all neoplasms.38 Similarly, practice patterns are reported in other countries.39 The widespread use of grading for bladder carcinoma reflects the success of the 1973 WHO grading system in patient management.

The main criticism of the 1973 WHO grading system is that patients with Grade 1 cancers are known to have a favorable prognosis. Therefore, some authorities question the validity of labeling these patients as having carcinoma. Very few patients with Grade 1 urothelial carcinoma will develop aggressive tumors or die of bladder carcinoma. The nomenclature PUNLMP was introduced in the recent bladder classification and grading systems1, 2 and has already encountered some resistance.40–47 Some investigators are concerned that the new classification merely represents changes in nomenclature, and that new descriptive terms (such as “PUNLMP”) may confuse clinicians.43 Furthermore, there is poor interobserver agreement for PUNLMP and low-grade urothelial carcinoma (κ = 0.12–0.50).41 After structured training for the new 1998 WHO/ISUP grading system, Murphy et al.41 concluded that neither refinements of morphologic criteria nor additional education would significantly decrease interpretive discrepancy in the grading of urothelial carcinoma. In a recent study, Bol et al.46 found that agreement among 3 experienced pathologists on the diagnosis of PUNLMP was 0%. Using the 1999 WHO classification system, the distribution of papilloma, PUNLMP, and Grade 1, Grade 2, and Grade 3 urothelial carcinoma was 0.8%, 0%, 50.8%, 25.4%, and 23%, respectively, after second review among 3 pathologists.46 Oyasu43 believed that the new term “PUNLMP” should not be used in the bladder classification considering that bladder carcinoma is a disease of field change48 and that progression to a high-grade tumor is a common event among patients who are constantly exposed to carcinogens.

In a large series of 112 patients diagnosed with PUNLMP, with a median follow-up of > 12 years, tumor recurrence was observed in 29% of the patients.7 Seventy-five percent of patients with tumor recurrence dedifferentiated into a higher tumor grade. The disease progression rate was 4%. Four patients developed invasive tumor and three died of bladder carcinoma.7 The tumor recurrence rate after the diagnosis of PUNLMP was reported to be 35% in the study by Holmang et al.49 and 47% in the study by Pich et al.50 In a study of 322 patients with a mean follow-up period of 79 months, Oosterhuis et al.40 found no difference with regard to tumor recurrence or disease progression between patients with PUNLMP and patients with low-grade urothelial carcinoma. They concluded that there are insufficient data to justify a different clinical approach or the introduction of a new pathologic category. Samaratunga et al.6 recently studied 134 patients with noninvasive papillary urothelial carcinoma from Johns Hopkins Hospital, and found that both the 1973 WHO and 1998 WHO/IUSP grading systems were predictive of patient outcome (P = 0.003 and P = 0.002, respectively). However, the disease progression rate was higher than in our previous studies.7, 33 With a median follow-up of 56 months, the 1973 WHO Grade 1 tumors were found to have a progression rate of 11%, whereas the 1998 WHO/IUSP PUNLMP tumors were found to have a progression rate of 8%.6 These data indicate that patients with PUNLMP have an increased risk of tumor recurrence, disease progression, and death from bladder carcinoma. Long-term clinical follow-up is recommended for these patients.

Recent advances in molecular genetics indicate that human bladder carcinoma arises from accumulated genetic alterations in oncogenes and tumor suppressor genes.8–20 Allelic loss of specific chromosome regions is an important aspect of disease progression. In the current study, we performed LOH analysis on 26 patients with the diagnosis of PUNLMP. Genomic DNA was extracted from formalin-fixed slides using laser capture microdissection. Eighty-one percent of patients with PUNLMP tumors had allelic loss in at least 1 of the DNA loci studied. The incidence of LOH in PUNLMP tumors was comparable to that found in urothelial carcinoma (Table 2).

Table 2. Comparison of LOH between PUNLMP and Urothelial Carcinoma
ReferencesStageGradeFrequency of LOH (%) Chromosome regions
9q32–33 D9S1779p22 IFNA17p13.1 TP5312q14–24 D12S10513p25–26 D3D3050
  1. LOH: loss of heterozygosity; PUNLMP: papillary urothelial neoplasm of low malignant potential.

Current studyTaPUNLMP4132292644
Paterson et al.18≥ T2G367 47 44
Louhelainen et al.19Ta–T1G2–3    33
Habuchi et al.15≥ T2G3    32
Knowles et al.10T1–T2    6 
Baud et al.20≥ T2G367    
Primdahl et al.13≥ T2G3 3541  
Uchida et al.51T1–4G2–3 3539  

In a previous study of muscle-invasive, high-grade urothelial carcinoma, the LOH rates at DNA loci D9S177, TP53, and D3S3050 were 67%, 47%, and 44%, respectively.18 Louhelainen et al.19 analyzed six patients with Ta–T1 Grade 2–3 urothelial carcinoma. In their study, the LOH rate at D3S3050 was 33%. Primdahl et al.13 studied 22 patients with muscle-invasive urothelial carcinoma using allelic deletion analysis. Of 17 informative patients, the authors found that allelic loss was 41% at locus TP53 and 35% at locus IFNA. In the current study, the rate of LOH was 41% with D9S177, 32% with IFNA, 29% with TP53, 26% with D12S1051, and 44% with D3S3050. Allelic loss of multiple chromosome loci was often observed in PUNLMP. The high frequency of LOH at various loci suggests that PUNLMP tumors share similar genetic instability with advanced stage bladder carcinoma.

Genetic changes that commonly occur in advanced stage bladder carcinoma are frequently found in PUNLMP tumors of the urinary bladder. Further studies are needed to ascertain whether such genetic changes influence the biologic behavior of PUNLMP, including risk of progression to more aggressive neoplasia, risk of tumor recurrence, and responsiveness to therapy.

Ancillary