Chromosomal abnormalities in renal cell carcinoma variants detected by Urovysion fluorescence in situ hybridization on paraffin-embedded tissue

Available online at www.sciencedirect.com

Annals of Diagnostic Pathology 15 (2011) 37 – 45

Chromosomal abnormalities in renal cell carcinoma variants detected by Urovysion fluorescence in situ hybridization on paraffin-embedded tissue Michelle D. Reid-Nicholson, MBBSa,⁎, Nisrin Motiwala, MBBSa , Scott C. Drury, MDa , Stephen C. Peiper, MDa , Martha K. Terris, MDb , Jennifer L. Waller, PhDc , Preetha Ramalingam, MBBSa b

Abstract

Keywords:

a Department of Pathology, Medical College of Georgia, Augusta, GA 30912, USA Department of Surgery, Section of Urology, Medical College of Georgia, Augusta, GA 30912, USA c Department of Biostatistics, Medical College of Georgia, Augusta, GA 30912, USA

Urovysion fluorescence in situ hybridization (UVFISH) identifies malignant cells in urine by detecting specific urothelial carcinoma-related chromosomal abnormalities. Some renal carcinomas (RCCs) share overlapping chromosomal aberrations with urothelial carcinoma. Malignant renal cells that are shed in urine can potentially cause a positive UVFISH result. We evaluated UVFISH in RCCs to determine its potential applicability to the diagnosis and grading of RCCs. Paraffin blocks from 39 RCCs (25 clear cell, 9 papillary, 2 chromophobe, and 3 sarcomatoid) and 15 controls (5 renal oncocytomas and 10 urothelial carcinomas) were tested. Of the RCCs, 15 (40%) were UVFISHpositive (9/25 [40%] clear cell, 3/9 [30%] papillary, 1/2 [50%] chromophobe, and 2/3 [67%] sarcomatoid carcinoma) and 24 (60%) were negative. Of the 15 controls, 8 (∼50%) were UVFISHpositive (2/5 [40%] oncocytomas and 6/10 [60%] urothelial carcinomas) and 7 (∼50%) were UVFISHnegative. Polysomy of chromosome 17 showed a statistically significant correlation with RCC subtype, being absent in most of the clear cell RCCs (P = .0096) compared with other RCCs. Polysomy of chromosome 7 was more frequent in high-grade than low-grade RCC (P = .0197) and more likely in high-grade clear cell than low-grade clear cell RCC (P = .0120). In conclusion, we showed that RCC has overlapping chromosomal abnormalities with urothelial carcinoma and can cause a positive UVFISH result. This has implications for the interpretation of Urovysion in patients whose urine contains malignant cells but who have negative cystoscopy and a concomitant renal mass. The chromosomal abnormalities observed in RCC are not distinct from those in urothelial carcinoma; therefore, UVFISH cannot distinguish these tumor types, nor can it type or grade RCC. © 2011 Elsevier Inc. All rights reserved. Urovysion FISH; Renal cell carcinoma

1. Introduction Urovysion fluorescence in situ hybridization (FISH) is a molecular-based assay that detects malignant urothelial cells in urine [1]. It was originally approved by the Federal Drug Administration for the surveillance of patients with estab⁎ Corresponding author. Department of Pathology, Emory University Hospital Midtown, Davis Fischer Building, Atlanta, GA 30308, USA. Tel.: +1 404 686 1995; fax: +1 404 686 4978. E-mail address: [email protected] (M.D. Reid-Nicholson). 1092-9134/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.anndiagpath.2010.07.011

lished bladder cancer; however, this was subsequently expanded to include its use in the initial diagnosis of bladder cancer in patients with subclinical disease [1-4]. Urovysion has a significantly higher sensitivity (73%-92%) and specificity (89%-96%) for urothelial carcinoma detection than urine cytology and cystoscopy [5]. Urovysion FISH probes are able to detect the most common chromosomal abnormalities associated with urothelial carcinoma. These include polysomy of chromosomes 3, 7, and 17, often seen in high-grade tumors, and loss of both 9p21 loci that is seen in most low-grade papillary tumors [6-8]. Although Urovysion

38

M.D. Reid-Nicholson et al. / Annals of Diagnostic Pathology 15 (2011) 37–45

was originally created for urine specimens, the test has been performed on paraffin-embedded tissues. It has also been adapted and used to evaluate malignant cells in effusions [9] and pancreatobiliary strictures [10]. To diagnose urothelial cancer, the Urovysion FISH assay relies on the identification of chromosomal abnormalities and not on cytomorphologic evaluation. Because the test cannot discriminate different histologic cell types, it is not surprising that non-urothelial genitourinary tract cancers sharing similar chromosomal aberrations with urothelial carcinoma would also cause a positive Urovysion FISH result. In a previous study we demonstrated this positivity in several non-urothelial bladder tumors including squamous cell carcinoma and primary and metastatic adenocarcinoma of the bladder [11]. Others have also shown similar results in small cell carcinoma of the genitourinary tract [11,12]. Another non-urothelial tumor that is known to involve the urinary tract is renal cell carcinoma (RCC). Malignant renal epithelium can penetrate the renal pelvis and enter the genitourinary tract, resulting in drop metastases to the ureter and bladder [13,14]. The presence of neoplastic renal epithelium in the urine of patients with renal carcinoma is also well documented in the cytology literature [15-20]. Some renal cancers share similar chromosomal abnormalities with urothelial carcinoma and hence could theoretically cause a positive Urovysion FISH result if these cells were shed in urine. We conducted a study to determine if RCC could cause a positive Urovysion FISH result in paraffinembedded tissue, and if so, could potentially be used to distinguish urothelial carcinoma from RCC, as well as to predict the subtype and the grade of RCC.

2. Materials and methods All RCCs diagnosed at the Medical College of Georgia between 2003 and 2008 were eligible for the study, which was approved by the institutional review board and complied with HIPAA regulations. Inclusion criteria were all RCCs whose histologic subtype and grade could be established by 2 reference pathologists and all cases with preserved paraffin blocks and an adequate number of neoplastic cells for analysis. Exclusion criteria were cases in which the carcinoma diagnosis, subtype, and grade could not be agreed upon by the 2 pathologists, cases without paraffin blocks, or cases with poorly preserved or limited tissue for analysis. Patient age, sex, tumor size, tumor type, grade, and stage were also documented. Twelve urothelial carcinomas and 5 renal oncocytomas were used as controls. Hematoxylin and eosin slides and immunohistochemical stains (when performed) from formalin-fixed, paraffin-embedded tissues were reviewed; and a diagnosis of RCC, renal oncocytoma, and urothelial carcinoma was rendered using World Health Organization (WHO) criteria. Only slides with more than 80% pure tumor

population and limited nontumor tissue were selected for analysis. An appropriate area on each slide was highlighted for hybridization. To ensure that the hybridization results would more accurately represent the tumors examined, only sections from resected tumors, and not biopsy samples, were used. Clear cell and chromophobe RCCs were graded using a modified version of the 4-tiered nuclear (Fuhrman) grading system [21]. The Fuhrman grading system constitutes the following: grade 1 tumors (at 10× objective) have small nuclei with inconspicuous nucleoli and resemble lymphocytes; grade 2 tumors have small nucleoli that cannot be visualized at 10× magnification; grade 3 tumor cells are slightly more pleomorphic and their nucleoli can be seen at 10× magnification; grade 4 tumors show obvious pleomorphism, hyperchromasia, and large nucleoli, and may also show sarcomatoid features. For the purpose of our study, we used a 2-tiered nuclear grading system lumping Fuhrman grade 1 and 2 renal cancers into a low-grade category and Fuhrman grade 3 and 4 cancers into a high-grade category. Although the Fuhrman grading system is also applicable to papillary RCC, for the purpose of the study, we subclassified these tumors into low grade and high grade, with low-grade denoting the type 1 tumors (papillae lined by a single layer of small cells with little cytoplasm) and high grade denoting the type 2 tumors (papillae lined by pseudostratified tumor cells of higher nuclear grade) [22]. All sarcomatoid RCCs (Fuhrman grade 4) were graded as high grade. The grading of urothelial carcinoma was based on current WHO guidelines [23]. 2.1. Fluorescence in situ hybridization The Urovysion probe mixture consisted of probes directed against the peri-centromeric regions of chromosomes 3 (CEP3), 7(CEP7), and 17 (CEP17), and the band 9p21 locus (LSI 9p21). Probes were labeled with Texas red (CEP3), spectrum green (CEP7), aqua (CEP17), and gold fluorophores (LSI 9p21). Paraffin sections were baked overnight at 56°C then deparaffinized through multiple rinses in Heme De. Slides were dehydrated for 5 min in 100% ethanol at room temperature, immersed in pretreatment solution at 80°C for 10 min, washed in purified water for 3 min, and then incubated in a protease buffer at 37°C for 15 min. They were then immersed in purified water for 3 min, followed by 10% formalin at room temperature for 10 min, washed in purified water, and then air dried. Successive washes with 70%, 85%, and 100% ethanol then followed. Probe solution was then added to the target area on the slides and a coverslip was placed. Slides and probes were co-denatured for 2 min at 75°C then placed overnight in a hybridization chamber at 37°C. After hybridization, unbound probes were washed and nuclei were counterstained with DAPI stain. A fluorescence microscope was used to visualize the red, green, gold, and aqua fluorescent Urovysion probe signals

M.D. Reid-Nicholson et al. / Annals of Diagnostic Pathology 15 (2011) 37–45

and DAPI counterstains using the following filters: DAPI single band pass, aqua single band pass (chromosome 17), gold single band pass (chromosome 9p21), and red/green dual band pass (chromosomes 3 and 7). Positive and negative controls were run concomitantly. A cell was abnormal if there were more than 2 red (chromosome 3), 2 green (chromosome 7), 2 aqua (chromosome 17) signals, or a loss of both copies of the gold (LSI 9p21) signals. A minimum of 25 tumor cells were evaluated for these chromosomal changes. If no abnormalities were detected, then the remaining cells were counted until a sufficient number of cells with chromosomal abnormalities were found or until 200 cells were evaluated. A positive result was ≥4 (or N10%) cells with more than 2 copies of chromosomes 3, 7, and 17. In the case of chromosome 9, a positive result was ≥12 cells showing zero 9p21 signals. Although the above criteria were established for the cytologic evaluation of urine samples, their usefulness in paraffin-embedded tissue has not been validated and cutoffs for abnormality in paraffin-embedded

39

Table 1 Summary of Urovysion FISH results by tumor type Tumor type

Positive FISH

Negative FISH

RCC (n = 39) Clear cell RCC (n = 25) ▪ Low grade ▪ High grade Papillary RCC (n = 9) ▪ Low grade ▪ High grade Chromophobe RCC (n = 2) ▪ Low grade ▪ High grade Sarcomatoid RCC (n = 3)

15/39 (40%) 9/25 (40%) 3 6 3/9 (30%) 3 0 1/2 (50%) 1 0 2/3 (67%)

24/39 (60%) 16/25 (60%) 11 5 6/9 (70%) 3 3 1/2 (50%) 0 1 1/3 (33%)

Controls (n = 15) Urothelial carcinoma (n = 10) ▪ Low grade ▪ High grade Oncocytoma (n = 5) Total no. of cases (n = 54)

8/15 (∼ 50%) 6/10 (60%) 0 6 2/5 (40%) 23/54 (40%)

7/15 (∼ 50%) 4/10 (40%) 1 3 3/5 (60%) 31/54 (60%)

RCC indicates renal cell carcinoma; n, number of cases.

Fig. 1. (A) Clear cell RCC, low grade (Fuhrman grade 1) (hematoxylin and eosin, original magnification ×200), with corresponding Urovysion FISH (green [chromosome 7], red [chromosome 3], gold [chromosome 9p21], and aqua [chromosome 17] probes) showing (B) polysomy of chromosome 3. (C) Clear cell RCC, high-grade (Fuhrman grade 3) (hematoxylin and eosin, original magnification ×400), with corresponding Urovysion FISH showing (D) polysomy of chromosomes 3, 7, 17, and loss of both 9p21 alleles.

40

M.D. Reid-Nicholson et al. / Annals of Diagnostic Pathology 15 (2011) 37–45

tissue where only pure populations of tumor cells are examined have not been established. 2.2. Statistical analysis Descriptive statistics were calculated on variables in cases and controls overall, and within all malignant renal tumor subtypes, as well as all urothelial carcinomas and oncocytomas. To examine the association between renal carcinoma subtypes and Urovysion FISH results, Fisher exact tests were performed because of the overall small sample size. Fisher exact tests were also used to determine the association of a positive or negative Urovysion FISH result (and specific chromosomal distribution) with each renal carcinoma subtype (chromophobe, clear cell, papillary, and sarcomatoid carcinomas combined) and then compared with controls (oncocytoma and urothelial carcinomas combined). The association of tumor grade in all renal carcinoma subtypes with FISH results and their specific chromosomal distribution was also examined. In all cases, statistical significance was assessed using an α level of .05.

3. Results A total of 58 tumors were selected for analysis including 41 renal cancers (27 clear cell renal carcinomas [15 low grade and 12 high grade], 2 chromophobe [1 low grade and 1 high grade], 9 papillary ([6 low grade and 3 high grade], and 3 sarcomatoid), 5 renal oncocytomas, and 12 urothelial carcinomas (2 low grade and 10 high grade). Thirty-eight percent of patients were female (n = 22) and 62% were male (n = 36). The mean age was 60.3 years (SD, 12.8 years). There was no statistically significant difference in patient sex (P = .8636) and tumor stage (P = .1743). Forty-six individuals had available data on tumor size, which averaged 5.59 cm (range, 1.1-15.0 cm; SD, 3.73 cm). Hybridization was successful in 54 of 58 cases and results are summarized in Table 1. The 4 tumors that did not hybridize included 2 clear cell carcinomas (1 low grade and 1 high grade) and 2 urothelial carcinomas (1 low grade and 1 high grade). Of the 39 RCCs that hybridized, 15 (40%) were FISHpositive and 24 (60%) were FISH-negative. The 25 clear cell carcinomas showed the following results: 9 (40%) of 25 were

Fig. 2. (A) Papillary RCC, low grade (type 1) (hematoxylin and eosin, original magnification ×100), with corresponding Urovysion FISH (green [chromosome 7], red [chromosome 3], gold [chromosome 9p21], and aqua [chromosome 17] probes) showing (B) polysomy of chromosomes 3, 7, and 17 and loss of both 9p21 alleles. (C) Oncocytoma (hematoxylin and eosin, original magnification ×200), with corresponding Urovysion FISH showing (D) polysomy of chromosome 3.

1.0000

.0022

.9410

.5960

60.0 40.0 80.0 20.0 80.0 20.0 100.0 0.0 100.0 0.0 Loss of 9p21

Polysomy 17

Polysomy 7

Polysomy 3

n indicates number of cases.

% n

3 2 4 1 4 1 5 0 5 0 40.0 60.0 60.0 40.0 70.0 30.0 50.0 50.0

% n

4 6 6 4 7 3 5 5 33.3 66.7 33.3 66.7 66.7 33.3 66.7 33.3 100.0 0.0

% n

1 2 1 2 2 1 2 1 3 0 0.0 100.0 0.0 100.0 100.0 0.0 0.0 100.0 100.0 0.0

% n

0 1 0 1 1 0 0 1 1 0 66.7 33.3 66.7 33.3 77.8 22.2 66.7 33.3 88.9 11.1

% n

6 3 6 3 7 2 6 3 8 1 Negative Positive Negative Positive Negative Positive Negative Positive Negative Positive FISH

%

65.4 34.6 69.2 30.8 80.8 19.2 96.1 3.9 92.3 7.7

n

Level

17 9 18 8 21 5 25 1 24 2

P Oncocytoma Urothelial carcinoma Sarcomatoid RCC Chromophobe RCC Papillary RCC Clear cell RCC

41

Variable

Table 2 Descriptive statistics and Fisher exact test results for all tumors combined

FISH-positive (3/9 [30%] were low grade and 6/9 [70%] were high grade); 16 (60%) of 25 of clear cell carcinomas were negative (11/16 [70%] were low grade and 5/16 [30%] were high grade). The papillary RCCs showed the following results: 3 (30%) of 9 were FISH-positive (all 3 [100%] were low grade) and 6 of 9 were negative (3/6 [50%] were low grade and 3/6 [50%] were high grade). One (50%) of 2 chromophobe RCCs was positive (high-grade tumor) and one, a low-grade tumor, was FISH-negative. Two (67%) of 3 sarcomatoid RCCs were FISH-positive and 1 (33%) of 3 was negative. Of the 15 controls, 8 (∼50%) were FISH-positive and 7 (∼50%) were FISH-negative (Table 1). Of the 5 oncocytomas, 2 (40%) were positive and 3 (60%) of 5 were negative. The urothelial carcinomas (n = 10) showed the following results: 6 (60%) of 10 were FISH-positive (all 6 were high grade) and 4 (40%) of 10 were negative for FISH (1 [25%] of the 4 was low grade and 3/4 [75%] were high grade). Figs. 1 and 2 show the renal tumors with positive Urovysion FISH results. Combining cases and controls, there was no statistically significant difference in the distribution of FISH results for all tumors combined (P = .5324). However, the distribution of polysomy of chromosome 17 showed a statistically significant difference (P = .0022) between tumors, with 96% of clear cell carcinomas and 100% of oncocytomas lacking polysomy of chromosome 17 compared with the other tumor types. The results of Fisher exact test for cases and controls combined, and for renal carcinoma subtypes (alone), are summarized in Tables 2 and 3. Among all positive renal cancers, polysomy of chromosome 3 was the most common (87%) followed by polysomy of chromosomes 17 (47%) and 7 (40%). Loss of both 9p21 loci was the least common abnormality (20%). Polysomy of chromosome 9p21 was observed in a single clear cell carcinoma but was not considered a positive FISH result. Three (100%) of 3 papillary RCCs showed polysomy of chromosomes 3 and 17, 2 (66%) of 3 showed polysomy of chromosome 7, and 1 (33%) of 3 showed loss of both 9p21 alleles. The most common chromosomal abnormality seen in sarcomatoid carcinoma was polysomy of chromosomes 3 and 17. The chromosomal abnormalities among all tumors with a positive FISH result are summarized in Table 4. On comparison of all 4 renal carcinoma subtypes, the only chromosomal abnormality that showed a statistically significant difference was polysomy of chromosome 17 (P = .0096), which was absent in 96% of FISH-positive clear cell carcinomas (compared with 67% of papillary and sarcomatoid carcinomas, and zero chromophobe carcinomas) (Table 3). None of the other chromosomal abnormalities showed a statistically significant difference between renal cancers. Among all controls (urothelial carcinoma and oncocytoma combined), polysomy of chromosomes 3 and 7 was seen in both tumors with oncocytomas showing 100% positivity for both abnormalities and urothelial carcinoma showing 50% and 33% positivity, respectively. Polysomy of

.5324

M.D. Reid-Nicholson et al. / Annals of Diagnostic Pathology 15 (2011) 37–45

42

M.D. Reid-Nicholson et al. / Annals of Diagnostic Pathology 15 (2011) 37–45

Table 3 Descriptive statistics and results of Fisher exact test for all renal carcinomas Variable

Gender Tumor grade Tumor stage

FISH Polysomy 3 Polysomy 7 Polysomy 17 Loss of 9p21

Level

Female Male High Low 1 2 3 4 Negative Positive Negative Positive Negative Positive Negative Positive Negative Positive

Chromophobe

Clear cell

Papillary

Sarcomatoid

n

%

n

%

n

%

n

%

1 1 1 1 2 0 0 0 0 1 0 1 1 0 0 1 1 0

50.0 50.0 50.0 50.0 100.0 0.0 0.0 0.0 0.0 100.0 0.0 100.0 100.0 0.0 0.0 100.0 100.0 0.0

11 16 13 14 15 2 9 1 17 9 18 8 21 5 25 1 24 2

40.7 59.3 48.2 51.8 55.6 7.4 33.3 3.7 65.4 34.6 69.2 30.8 80.8 19.2 96.1 3.9 92.3 7.7

3 6 3 6 8 0 1 0 6 3 6 3 7 2 6 3 8 1

33.3 66.7 33.3 66.7 88.9 0.0 11.1 0.0 66.7 33.3 66.7 33.3 77.8 22.2 66.7 33.3 88.9 11.1

2 1 3 0 0 0 3 0 1 2 1 2 2 1 2 1 3 0

66.7 33.3 100.0 0.0 0.0 0.0 100.0 0.0 33.3 66.7 33.3 66.7 66.7 33.3 66.7 33.3 100.0 0.0

P

.8772 .2966 .1743

.4164 .3628 .8653 .0096 1.0000

n, number of cases.

chromosome 17 was present in 3 (50%) of 6 urothelial carcinomas but not in oncocytomas (Table 3). Table 5 shows the results of Fisher exact test by case/control status. Regarding tumor stage in RCC, there was no statistically significant difference between the distribution of a positive FISH result and the stage of renal cancer (when all subtypes were combined, and when clear cell carcinoma and papillary

carcinoma were examined individually). However, regarding tumor grade in the renal cancers (both combined, and among the clear cell and papillary carcinomas individually), there was a statistically significant difference in the distribution of polysomy of chromosome 7 (P = .0197). Low-grade tumors were less likely to have polysomy of chromosome 7 (only 10% of these tumors had polysomy 7), whereas 37% of high-

Table 4 Chromosomal abnormalities in all tumors with positive Urovysion FISH results Tumor type

Polysomy 3

Polysomy 17

Loss of 9p21

Polysomy 9p21a

0 2 2/9 (22%)

0 2 2/9 (22%)

0 1 1/9 (11%)

3 3/3 (100%)

1 1/3 (33%)

0 0

0 0 2 2/2 (100%) 7/15 (47%)

0 0 0 0 3/15 (20%)

0 0 0 0 1/15 (b1%)

2 2/2 (100%)

0 0

0 0

0 0

2 2/6 (33%) 4/8 (50%)

3 3/6 (50%) 3/8 (40%)

0 0 0

0 0 0

Polysomy 7

Renal carcinomas with positive Urovysion FISH results (n = 15) Clear cell RCC (n = 9) Low grade (n = 3) 3 0 High grade (n = 6) 4 3 Total 7/9 (78%) 3/9 (33%) Papillary RCC (n = 3) Low grade (n = 3) 3 2 Total 3/3 (100%) 2/3 (66%) Chromophobe RCC (n = 1) High grade (n = 1) 1 0 Total 1/1 (100%) 0 Sarcomatoid RCC (n = 2) 2 1 Total 2/2 (100%) 1/2 (50%) Total renal cancers (n = 15) 13/15 (87%) 6/15 (40%) Controls with positive Urovysion FISH results (n = 8) Oncocytoma (n = 2) 2 Total 2/2 (100%) Urothelial carcinoma (n = 6) High grade 3 Total 3/6 (50%) Total no. controls (n = 8) 5/8 (60%)

n indicates number of cases. a Polysomy of chromosome 9p21 is not a feature of a positive FISH result but is included in the table because it was observed in a single case of clear cell RCC.

M.D. Reid-Nicholson et al. / Annals of Diagnostic Pathology 15 (2011) 37–45

43

Table 5 Descriptive statistics and Fisher exact test results by case/control status

4. Discussion

Variable

Since the early 1959s, numerous studies have documented the presence of neoplastic renal epithelium in urine [15-20,24]. This phenomenon has been described in clear cell RCC [15,16,18], papillary RCC [17], and collecting duct carcinoma [19]. Of the various types of renal cancers, highgrade clear cell carcinoma, high-grade papillary carcinoma, and collecting duct carcinoma are the most likely to be shed in urine. This is because these tumors are the most aggressive, and most locally invasive, and hence have an increased propensity for penetrating the renal pelvis [20]. Renal carcinoma is associated with multiple chromosomal abnormalities, including deletions of chromosome 3p (clear cell carcinoma [25], collecting duct carcinoma [26]), deletions of chromosomes 1, 2, and 6 (chromophobe carcinoma [27]), and polysomy of chromosomes 7 and 17 (papillary RCC [28]). Because some of these chromosomal abnormalities are similar to those seen in urothelial carcinoma, if these neoplastic renal cells were shed in urine they could potentially cause a positive FISH result. If Urovysion could reliably and consistently identify these malignant renal cells in urine samples, then one could potentially exploit the test for the screening and diagnosis of RCC in patients with subclinical disease. Renal carcinoma is the 12th most common carcinoma worldwide, and the number of new cases diagnosed annually has been steadily increasing [29]. In 2009, in the United States alone, there will be 49,096 new cases of kidney cancer and 11,033 related deaths [30]. Because most renal cancers are clinically silent, more than 30% of patients present with metastatic disease [31]. If the Urovysion kit could detect tumor cells in the urine of these “silent” cases, it would facilitate detection of subclinical disease. We tested this hypothesis in several types of renal cancers, with an aim at identifying distinguishing chromosomal abnormalities that could potentially be used to diagnose, type, and grade RCC, as well as distinguish it from urothelial carcinoma. We showed that RCC can cause a positive Urovysion FISH result, in as many as 40% of the cases that we tested. This would imply that if RCC breaches the renal pelvis and enters the urinary tract, the likelihood of a positive FISH result is significantly increased. Although we examined several different types of RCC, our study focused primarily on clear cell and papillary tumors because these are the most common and are the most likely to be shed in urine. Positive FISH results showed a similar distribution across the 4 RCC subtypes examined. There was no one chromosomal abnormality that achieved statistical significance in terms of its ability to diagnose and type RCC, or to differentiate it from urothelial carcinoma. Although polysomy of chromosome 3 was the most common chromosomal abnormality in the renal cancers, being twice as common as polysomy of chromosome 7 or 17, this result did not achieve statistical significance. We also found that 78% of clear cell RCCs showed polysomy of chromosome 3. This is not a typical

FISH Polysomy 3 Polysomy 7 Polysomy 17 Loss of 9p21

Level

Negative Positive Negative Positive Negative Positive Negative Positive Negative Positive

Cases (n = 39)

Controls (n = 15)

P

n

%

n

%

24 15 25 14 31 8 33 6 36 3

61.5 30.5 64.1 35.9 79.5 20.5 84.6 15.4 92.3 7.7

7 8 10 5 11 4 10 5 5 0

46.7 53.3 66.7 33.3 73.3 26.7 66.7 33.3 100.0 0.0

.3689 1.0000 .7189 .2558 1.0000

n indicates number of cases.

grade tumors were more likely to have polysomy of chromosome 7 (Table 6). Polysomy of chromosome 7 was also significantly more likely (P = .0120) in high-grade clear cell carcinoma (42% of cases) than in low-grade clear cell carcinoma (0 cases had polysomy of chromosome 7). Table 6 Results of Fisher exact test by tumor grade for all renal carcinomas Variable

High grade

Low grade

n

n

%

50.0 50.0 58.3 41.7 58.3 41.7 91.7 8.3 83.3 16.7

11 3 11 3 14 0 14 0 14 0

78.6 21.4 78.6 21.4 100.0 0.0 100.0 0.0 100.0 0.0

Negative 3 100.0 Positive 0 0.0 Polysomy 3 Negative 3 100.0 Positive 0 0.0 Polysomy 7 Negative 2 66.7 Positive 1 33.3 Polysomy 17 Negative 2 66.7 Positive 1 33.3 Loss of 9p21 Negative 3 100.0 Positive 0 0.0 All renal carcinomas regardless of subtype FISH Negative 10 52.6 Positive 9 47.4 Polysomy 3 Negative 11 57.9 Positive 8 42.1 Polysomy 7 Negative 12 63.2 Positive 7 36.8 Polysomy 17 Negative 15 79.0 Positive 4 21.0 Loss of 9p21 Negative 17 89.5 Positive 2 10.5

3 3 3 3 5 1 4 2 5 1

50.0 50.0 50.0 50.0 83.3 16.7 66.7 33.3 83.3 16.7

14 6 14 6 18 2 23 2 19 1

70.0 30.0 70.0 30.0 90.0 10.0 92.0 8.0 95.0 5.0

Clear cell RCC FISH Polysomy 3 Polysomy 7 Polysomy 17 Loss of 9p21

Level

Negative Positive Negative Positive Negative Positive Negative Positive Negative Positive

Papillary RCC FISH

n indicates number of cases.

6 6 7 5 7 5 11 1 10 2

%

P

.2177 .4009 .0120 .4615 .2031

.4643 .4643 .0000 1.0000 1.0000

.3332 .5145 .0197 .4075 .6050

44

M.D. Reid-Nicholson et al. / Annals of Diagnostic Pathology 15 (2011) 37–45

finding in clear cell carcinoma, and we cannot explain this observation in our series. Other chromosomal abnormalities that are commonly seen in clear cell RCC such as monosomy and deletions of chromosomes 4p, 14q, 9p, and Y [32] are not tested for by Urovysion. We also noted that papillary RCC showed polysomy of chromosomes 7 and 17. This is similar to the observations of others who have shown that the most common chromosomal abnormalities in papillary RCC are trisomy of chromosomes 7 and 17, as well as trisomy of chromosome 20 and loss of the Y chromosome [28,32]. More gains of chromosomes 7p and 17p are seen in type 1 papillary RCC than type 2 tumors [22]. Type 1 tumors are also associated with lower stage disease and longer survival than type 2 tumors [33]. We examined 9 papillary carcinomas and when we compared the type 1 and type 2 tumors, there was no significant difference in chromosomal abnormalities by Urovysion FISH. However, we only examined a small number of papillary carcinomas. The 2 positive sarcomatoid RCCs had polysomy of chromosomes 3, 7, and 17, none of which have documented specificity in this variant of carcinoma. One chromophobe carcinoma had a positive Urovysion FISH result and showed polysomy of chromosome 3. Of all the renal cancers we examined, chromophobe carcinoma is perhaps the least likely to be shed in urine because of its more prototypic low nuclear grade and early clinical stage at diagnosis. Chromophobe RCC tends to show significant chromosomal loss including loss of chromosome 17, which would not be interpreted as a positive FISH result [27]. Although we did not examine any cases of collecting duct carcinoma, the shedding of these neoplastic cells in urine has been documented by others [19]. Collecting duct carcinoma is a high-grade (Fuhrman 3 or 4) tumor with a propensity for aggressive behavior and as such it can invade the renal pelvis and can be shed in urine. The chromosomal abnormalities most frequently described in collecting duct carcinoma include loss of chromosome 3p and monosomy of chromosomes 1, 6, 14, 15, and 22, but none of these would lead to a positive FISH result [26]. We found that renal oncocytomas also caused a positive FISH result, showing polysomy of chromosomes 3 and 7. They also lacked polysomy of chromosome 17. Several articles describe rare chromosomal abnormalities in renal oncocytoma, including loss of chromosomes 1 and 14, and translocation of t(5;11), but polysomy of chromosome 3 has to our knowledge not been previously reported in renal oncocytoma [34]. Although the latter observation was an interesting and unexpected one, it has little clinical significance in the context of this study as oncocytomas are benign renal tumors and are unlikely to be shed in urine. The findings of our study have implications for the dayto-day practice of cytopathology. A patient whose urine has malignant epithelial cells, a positive Urovysion assay, negative cystoscopic findings, and a concomitant renal mass can create a diagnostic challenge for cytopathologists. If an inappropriate emphasis is placed on FISH results alone, diagnostic errors can be made that have far-reaching

implications for patient management and outcome. Distinguishing urothelial carcinoma from RCC on cytomorphology alone can be extremely difficult in urine. When these malignant cells cause a positive FISH result and are associated with a negative cystoscopy or intravenous pyelogram, the final pathologic diagnosis may ultimately rest on histologic examination of the resected tumor. In conclusion, our data confirm that RCC shares overlapping chromosomal abnormalities with urothelial carcinoma; and if malignant renal cells breach the renal pelvis and enter the urinary tract, they could theoretically cause a positive Urovysion FISH. Cytopathologists and urologists should be mindful of this potential pitfall and cautiously interpret positive FISH results, especially in patients with renal masses and negative cystoscopic findings. Because the chromosomal abnormalities in RCC are not distinct from those in urothelial carcinoma, Urovysion FISH cannot be used to distinguish these 2 tumors, nor can it be used to type or grade RCC. Our findings open the door for others to develop newer broad-spectrum FISH assays capable of identifying the most consistent and predictable chromosomal abnormalities in RCC. Acknowledgments The authors thank Mrs Kimberly Smith and Mrs Karen Norris for their expert technical assistance in the completion of this study. References [1] Halling KC. Vysis UroVysion for the detection of urothelial carcinoma. Expert Rev Mol Diagn 2003;3(4):507-19. [2] Skacel M, Fahmy M, Brainard JA, et al. Multitarget fluorescence in situ hybridization assay detects transitional cell carcinoma in the majority of patients with bladder cancer and atypical or negative urine cytology. J Urol 2003;169(6):2101-5. [3] Sokolova IA, Halling KC, Jenkins RB, et al. The development of a multitarget, multicolor fluorescence in situ hybridization assay for the detection of urothelial carcinoma in urine. J Mol Diagn 2000;2(3): 116-23. [4] US Food and Drug Administration Center for Devices and Radiological Health. New device approval: Urovysion bladder cancer: P030052. January 24. Vol. Available at: http://www.fda.gov/cdrh/mda/docs/ p030052.html. [5] Halling KC, King W, Sokolova IA, et al. A comparison of cytology and fluorescence in situ hybridization for the detection of urothelial carcinoma. J Urol 2000;164(5):1768-75. [6] Mitra AP, Datar RH, Cote RJ. Molecular pathways in invasive bladder cancer: new insights into mechanisms, progression, and target identification. J Clin Oncol 2006;24(35):5552-64. [7] Bergman J, Reznichek RC, Rajfer J. Surveillance of patients with bladder carcinoma using fluorescent in-situ hybridization on bladder washings. BJU Int 2008;101(1):26-9. [8] Sugano K, Kakizoe T. Genetic alterations in bladder cancer and their clinical applications in molecular tumor staging. Nat Clin Pract Urol 2006;3(12):642-52. [9] Flores-Staino C, Darai-Ramqvist E, Dobra K, Hjerpe A. Adaptation of a commercial fluorescent in situ hybridization test to the diagnosis of malignant cells in effusions. Lung Cancer 2010;68(1):39-43.

M.D. Reid-Nicholson et al. / Annals of Diagnostic Pathology 15 (2011) 37–45 [10] Fritcher EG, Kipp BR, Halling KC, et al. A multivariable model using advanced cytologic methods for the evaluation of indeterminate pancreatobiliary strictures. Gastroenterology 2009;136(7):2180-6. [11] Reid-Nicholson MD, Ramalingam P, Adeagbo B, Cheng N, Peiper SC, Terris MK. The use of Urovysion fluorescence in situ hybridization in the diagnosis and surveillance of non-urothelial carcinoma of the bladder. Mod Pathol 2009;22(1):119-27. [12] Yoder BJ, Skacel M, Hedgepeth R, et al. Reflex UroVysion testing of bladder cancer surveillance patients with equivocal or negative urine cytology: a prospective study with focus on the natural history of anticipatory positive findings. Am J Clin Pathol 2007;127(2):295-301. [13] Mayer WA, Resnick MJ, Canter D, et al. Synchronous metastatic renal cell carcinoma to the genitourinary tract: two rare case reports and a review of the literature. Can J Urol 2009;16(2):4611-4. [14] Sim SJ, Ro JY, Ordonez NG, Park YW, Kee KH, Ayala AG. Metastatic renal cell carcinoma to the bladder: a clinicopathologic and immunohistochemical study. Mod Pathol 1999;12(4):351-5. [15] Foot NC, Papanicolaou GN, Holmquist ND, Seybolt JF. Exfoliative cytology of urinary sediments; a review of 2,829 cases. Cancer 1958; 11(1):127-37. [16] Hajdu SI, Savino A, Hajdu EO, Koss LG. Cytologic diagnosis of renal cell carcinoma with the aid of fat stain. Acta Cytol 1971;15(1):31-3. [17] Kawakami H, Hoshida Y, Hanai J, et al. Voided urine cytology of papillary renal cell carcinoma and renal calculus: report of a case with emphasis on the importance of cytologic screening in high-risk individuals. Acta Cytol 2001;45(5):771-4. [18] Piscioli F, Detassis C, Polla E, Pusiol T, Reich A, Luciani L. Cytologic presentation of renal adenocarcinoma in urinary sediment. Acta Cytol 1983;27(4):383-90. [19] Nguyen GK, Schumann GB. Cytopathology of renal collecting duct carcinoma in urine sediment. Diagn Cytopathol 1997;16(5):446-9. [20] Umiker W. Accuracy of cytologic diagnosis of cancer of the urinary tract. Symposium on Diagnostic Accuracy of Cytologic Technics 1964;8(3):186-91. [21] Fuhrman SA, Lasky LC, Limas C. Prognostic significance of morphologic parameters in renal cell carcinoma. Am J Surg Pathol 1982;6(7):655-63. [22] Delahunt B, Sauter G, Epstein JI, Sesterhenn IA, editors. World Health Organization classification of tumours. Pathology and genetics of tumours of the urinary system and male genital system. 1st ed. Lyon, France: IARC Press; 2004. p. 27-9.

45

[23] Gasser T, Hartmann A, Schmitz-Drager BJ, et al. Infiltrating urothelial carcinoma. In: Eble JN, Sauter G, Epstein JI, Sesterhenn IA, editors. World Health Organization classification of tumours. Pathology and genetics of tumours of the urinary system and male genital system. 1st ed. Lyon, France: IARC Press; 2004. p. 93-109. [24] Harrison JH, Botsford TW, Tucker MR. The use of the smear of the urinary sediment in the diagnosis and management of neoplasm of the kidney and bladder. Surg Gynecol Obstet 1951;92(2):129-39. [25] Bonsib SM, Moch H. Clear cell renal cell carcinoma. In: Eble JN, Sauter G, Epstein JI, Sesterhenn IA, editors. World Health Organization Classification of tumours. Pathology and genetics of tumours of the urinary system and male genital system. 1st ed. Lyon, France: IARC Press; 2004. p. 23-5. [26] Fuzesi L, Cober M, Mittermayer C. Collecting duct carcinoma: cytogenetic characterization. Histopathology 1992;21(2):155-60. [27] van den Berg E. Chromophobe renal cell carcinoma. In: Eble JN, Sauter G, Epstein JI, Sesterhenn IA, editors. World Health Organization classification of tumours. pathology and genetics of tumours of the urinary system and male genital system. 1st ed. Lyon, France: IARC Press; 2004. p. 30-2. [28] Kovacs G, Fuzesi L, Emanual A, Kung HF. Cytogenetics of papillary renal cell tumors. Genes Chromosomes Cancer 1991;3(4):249-55. [29] Eble JNTK, Pisani P. Renal cell carcinoma. In: Eble JN, Sauter G, Epstein JI, Sesterhenn IA, editors. World Health Organization classification of tumours. Pathology and genetics of tumours of the urinary system and male genital system. 1st ed. Lyon, France: IARC Press; 2004. [30] http://www.cancernet.gov/cancertopics/types/kidney. [31] Skinner DG, Colvin RB, Vermillion CD, Pfister RC, Leadbetter WF. Diagnosis and management of renal cell carcinoma. A clinical and pathologic study of 309 cases. Cancer 1971;28(5):1165-77. [32] Klatte T, Rao PN, de Martino M, et al. Cytogenetic profile predicts prognosis of patients with clear cell renal cell carcinoma. J Clin Oncol 2009;27(5):746-53. [33] Klatte T, Pantuck AJ, Said JW, et al. Cytogenetic and molecular tumor profiling for type 1 and type 2 papillary renal cell carcinoma. Clin Cancer Res 2009;15(4):1162-9. [34] Moch H. Oncocytoma. In: Eble JN, Sauter G, Epstein JI, Sesterhenn IA, editors. World Health Organization classification of tumours. Pathology and genetics of tumours of the urinary system and male genital system. 1st ed. Lyon, France: IARC Press; 2004. p. 42-3.