Diagnosis of upper tract urothelial carcinoma—a comparative study of urinary cytology and surgical biopsy

Diagnosis of upper tract urothelial carcinoma—a comparative study of urinary cytology and surgical biopsy

Journal of the American Society of Cytopathology (2015) 4, 3e9 Available online at www.sciencedirect.com ScienceDirect journal homepage: www.jascyto...

532KB Sizes 0 Downloads 133 Views

Journal of the American Society of Cytopathology (2015) 4, 3e9

Available online at www.sciencedirect.com

ScienceDirect journal homepage: www.jascyto.org/

ORIGINAL ARTICLE

Diagnosis of upper tract urothelial carcinomada comparative study of urinary cytology and surgical biopsy Lu Wang, MD, Stefan E. Pambuccian, MD, Eva M. Wojcik, MD, Güliz A. Barkan, MD* Department of Pathology and Lab Medicine, Loyola University Medical Center, 2160 W. 1st Avenue, Maywood, Illinois Received 10 January 2014; received in revised form 2 September 2014; accepted 3 September 2014

KEYWORDS upper tract; urothelial carcinoma; urinary cytology; biopsy; sensitivity

Introduction Upper tract urothelial carcinoma (UTUC) is defined as urothelial carcinoma (UC) arising in the renal pelvis and ureter. Upper tract (UT) cytology and biopsy evaluation can be technically challenging. The aim of the study is to evaluate the diagnostic modality and sensitivity of urine cytology and ureteroscopic biopsy for the diagnosis of UTUC. Material and methods All patients with UTUC who underwent radical nephroureterectomy or ureterectomy with preoperative cytology and/or biopsy from January 1, 2000 to September 30, 2011 at our institution were included in this study. The sensitivity of each diagnostic modality was calculated with respect to tumor grade, stage, and size. Results A total of 143 cytology specimens and 54 biopsies from 65 patients were evaluated. For low-grade UTUC, the sensitivities for biopsy, lower tract cytology, and UT cytology were 68.4%, 27.3%, and 37.5%, respectively. These numbers were 82.9%, 40.7%, and 80.6% for high-grade UTUC. By combining the UT cytology and biopsy, the diagnostic sensitivities were increased to 87.5% for low-grade UTUC and 100% for high-grade UTUC. The consistency of tumor grade between biopsy and surgical specimen were 63.2% for low-grade UTUC and 68.6% for highgrade UTUC. Conclusions Both UT cytology and biopsy showed higher sensitivity in detecting high-grade UTUC versus low-grade UTUC. The sensitivities of UT cytology and ureteroscopic biopsy in detecting high-grade UTUC were comparable. The sensitivity was greatly improved when these diagnostic modalities were combined. As expected, the selective UT cytology evaluation had superior sensitivity in detecting UTUC than did the lower tract cytology sampling. Ó 2015 American Society of Cytopathology. Published by Elsevier Inc. All rights reserved.

*Corresponding author: Güliz A. Barkan, MD, Department of Pathology and Lab Medicine, Loyola University Medical Center, 2160 W. 1st Avenue, Maywood, IL 60153; Tel.: (708) 327-2572; Fax: (708) 327-2620. E-mail address: [email protected] (G.A. Barkan). 2213-2945/$36 Ó 2015 American Society of Cytopathology. Published by Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jasc.2014.09.203

4

Introduction Accurate preoperative diagnosis of upper tract urothelial carcinoma (UTUC) is essential for formulating an appropriate therapeutic strategy. Although imaging has a very important contribution,1,2 urinary tract cytology and biopsy diagnosis still play a crucial role in UTUC management.3,4 However, adequate ureteroscopic biopsies are difficult to obtain and may lead to diagnostic difficulties as shown by Tavora et al.4 who concluded that “the pathologists should recognize that in almost one of the four renal pelvic/ureteral biopsies, a definitive diagnosis cannot be made because of the inadequate tissue.” As a consequence of technical difficulties in obtaining adequate biopsies of the upper urinary tract,4 cytology plays an essential role in the diagnosis of UTUC. The sensitivity of urinary cytology in detecting urothelial carcinomas (UC) of all sites has been well-studied and ranges between 55% and 90%.5-8 Only a few upper tract (UT) cytology (ie, selective ureteral/renal pelvis cytology) studies on UTUC have been reported, with similar, albeit lower sensitivities, ranging between 43% and 78%.3,9-11 The aim of this study was to better understand the diagnostic value of urinary cytology and ureteroscopic biopsy in detecting UTUC. To achieve this goal, we performed a retrospective cohort study of patients with UTUC managed at our institution, and we determined the sensitivities of urinary cytology and ureteroscopic biopsy in detecting UTUC, as well as the impact of tumor grade, size, and stage on the sensitivities of these diagnostic methods.

Material and methods We performed a search of the electronic medical records to identify patients who were diagnosed with UTUC and subsequently underwent a radical nephroureterectomy or ureterectomy at our institution between January 1, 2000 and September 30, 2011. Patients who had no urinary cytology or ureteroscopic biopsy performed within a 6-month interval prior to surgery were excluded from the study. Patients with previously or concomitantly diagnosed UC of the bladder were excluded in order to allow for comparison for UT cytology with bladder barbotage specimens. The tumors were staged according to the 2002 American Joint Committee on Cancer TNM (tumor, node, metastasis) classification,12 and graded according to the 2004 World Health Organization/1998 International Society of Urologic Pathology consensus classification.13-15 At our institution, UT cytology cases are categorized in a 4-tier system: positive for UC (positive for high grade UC [HGUC]); suspicious for UC (suspicious for HGUC); atypical urothelial cells present (AUC); and negative (negative for malignancy). On the rare occasion when a true fibrovascular core is visualized and there is minimal cytologic atypia, a diagnosis of low-grade UC (LGUC) is rendered. The diagnosis “suspicious for HGUC” is only rendered when there are rare

L. Wang et al. cells (<10 cells) with the cytomorphologic features of HGUC; this category is analogous to the recently proposed AUC-H (atypical urothelial cells, cannot exclude high-grade urothelial carcinoma).2 At most institutions, including ours, the urological management of these “suspicious” cases is the same as that of the “positive for HGUC” cases. As such, for the purposes of this study, these groups were combined when calculating the sensitivity of urinary cytology. Due to the uncertainty of the exact meaning and management of an “atypical” (AUC) diagnosis, analysis was done both considering AUC diagnoses as “positive” and as “negative” for sensitivity calculations. The urinary cytology evaluation was further subgrouped into upper tract (selective ureteral/ renal pelvis washing or brushing) and lower tract ([LT], ie, voided urine and bladder barbotage) cytology. Two-tailed P values were calculated by the Fischer exact test, based on 2  2 contingency tables. A P value < 0.05 was set for statistical significance. As the patient pool consisted of patients with confirmed UTUC alone, only sensitivity data could be calculated.

Results We identified 80 patients who underwent radical nephroureterectomy or ureterectomy for UTUC from January 2000 to September 2011. Fifteen patients were excluded because they either did not have any cytologic specimens (n Z 10) or did not have “adequate” preoperative urinary cytology or biopsies, including 2 biopsy specimens in which the “tissue” was lost during processing and 3 cytology cases in which the specimens were completely acellular and therefore “unsatisfactory” for cytologic evaluation. The study group consisted of 65 patients, 36 men ages 37 to 83 years (mean 68) and 29 women ages 51 to 89 years (mean 72), who had preoperative cytology (143 cytologic specimens from 59 patients), and ureteroscopic biopsies (54 biopsies from 53 patients). Of those 65 patients, 46 underwent testing by both modalities. From the 54 preoperative biopsy specimens, 35 were from the renal pelvis and 19 from the ureters. Of the 143 urine cytology specimens, 59 were from the renal pelvis, 35 from the ureters, and 49 were bladder barbotage or voided urine specimens (Table 1). The preoperative cytology results showed 79 cases that were positive or suspicious for UC (Fig. 1 depicts an example of HGUC), of which 5 cases (all renal pelvic brushing specimens) were reported as LGUC. An additional 25 cases were reported as AUC, and 39 cases were reported as negative for malignancy. In none of the cases was LT cytology positive or suspicious for UC when UT cytology was negative. UT cytology had a significantly higher sensitivity than LT cytology did, independent of the assignment of AUC cases to the positive or negative categories (62 of 94 [66.0%] versus 17 of 49 [34.7%], P Z 0.0004 if AUC was considered negative and

Comparative study of upper tract urinary cytology and surgical biopsy Table 1

5

Comparison of the sensitivities of urinary cytology and ureteroscopic biopsy disregarding tumor grade.

Diagnostic method

Positivea

AU

False negative

Total

Sensitivity when AU excluded/included

Ureteroscopic biopsy Urinary cytology

42 79 17 62

2 25 11 14

10 39 21 18

54 143 49 94

77.8%/81.5% 55.2%/72.7% 34.7%/57.1% 66.0%/80.9%

Combined Lower tract Upper tract

Abbreviation: AU, atypical urothelial cells present. a Positive results include “positive for urothelial carcinoma” and “suspicious for urothelial carcinoma”.

76 of 94 [80.9%] versus 28 of 49 [57.1%], P Z 0.005 when AUC were considered positive). The sensitivity of UT cytology for detecting HGUC was significantly higher than that of low-grade UC (50 of 62 [80.6%] versus 12 of 32 [37.5%], P < 0.0001 if AUC was considered negative and 57 of 62 [91.9%] versus 19 of 32 [59.4%], P Z 0.0005, when AUC were considered positive). The sensitivity of biopsies was higher for detecting HGUC than for low-grade UC (29 of 35 [82.9%] versus 13 of 19 [68.4%]), but the difference was not statistically significant (P Z 0.310) (Table 2). UT cytology showed better sensitivity for HGUC (but not for LGUC) than LT cytology. Compared with UT cytology, LT cytology had lower sensitivity for detecting HGUC (40.7%) (P < 0.001), but no significant difference for detecting LGUC (27.3%) (P Z 0.433). In 46 patients who had both preoperative biopsy and cytology, combining the results of these 2 modalities increased the sensitivity to 100%. Nineteen of 65 patients had concomitant UT washing and brushing specimens (12 renal pelvis, 7 ureter). All 19 brushing specimens were diagnosed as “positive for UC.” However, 2 of 19 of the washing cases were diagnosed as “negative.” The 2 “negative” cases were reviewed again

microscopically and only benign urothelial cells were identified. The nephroureterectomy on these cases showed a LGUC. Hence, overall, the concordance between UT washing and brushing was 89.5%. The grading of the ureteroscopic biopsy was also compared with the final surgical diagnosis. The concordance was 63.2% for LGUC and 68.6% for HGUC, unlike the perfect correlation reported by previous studies.15 Overall, neither cytology nor biopsy sensitivities showed any trend when correlated with the tumor size. However, as expected, LGUC correlated well with the lower tumor stages (Ta, Tis, and T1). On the other hand, most HGUC (54.2% in positive cytology specimens and 52% in positive biopsy specimens) were staged as T3 (Tables 3 and 4). When HGUC were grouped according the pT stage as a low-stage group (Ta, Tis, and T1) and high-stage group (T2 or above), the sensitivities of urine cytology were 66.7% (20 of 30) and 91.2% (52 of 57) for low-stage and high-stage carcinomas, respectively; P Z 0.015. Similarly, sensitivities of UT biopsies for each group were 77.8% (7 of 9) and 85.7% (18 of 21) for low-stage and high-stage carcinomas respectively; P Z 0.593.

Figure 1 High-grade urothelial carcinoma. Urine cytology specimen shows high-grade urothelial carcinoma cells with enlarged nuclei, nuclear hyperchromasia, irregular nuclear contour, and high nuclear-to-cytoplasmic ratio (>0.5) (A). Corresponding resection specimen shows multiple layers of disorganized tumor cells with enlarged nuclei, clumpy chromatin, and increased nuclear-to-cytoplasmic ratio (B).

6

L. Wang et al. Table 2

Comparison of the sensitivities of the biopsy, cytology, and coevaluation according to tumor grade.

Tumor grade Low-grade UTUC

High-grade UTUC

Results

Biopsy

Positivea AU Negative Total Sensitivity (AU was considered negative/positive) Positivea AU Negative Total Sensitivity (AU was considered negative/positive)

Cytology Lower tract

Upper tract

13 1 5 19 68.4%/73.7%

6 5 11 22 27.3%/50.0%

12 7 13 32 37.5%/59.4%

29 1 5 35 82.9%/85.7%

11 6 10 27 40.7%/63.0%

50 7 5 62 80.6%/91.9%

Coevaluation of biopsy and cytology 14 2 16 87.5%

30 0 30 100%

Abbreviations: AU, atypical urothelial cells present; UTUC, upper tract urothelial carcinoma. a Positive results include “positive for urothelial carcinoma” and “suspicious for urothelial carcinoma”.

regard the diagnosis of “atypical urothelial cells present” as a negative diagnosis. However, this may not be the universal approach. Therefore, we present sensitivity calculations in both ways by including and excluding the AUC cases. The value of UT endoscopic biopsy has not been adequately studied, but it appears to be comparable to that of cytology, with reported accuracy rates of UT biopsy ranging between 43% and 89%.2,3,9 Tavora et al.4 retrospectively reviewed 76 consecutive mid-upper ureter and renal pelvis biopsies and found 7 cases, which were initially diagnosed as an urothelial neoplasm, yet were nonneoplastic on review. Strips of urothelium without welldeveloped fibrovascular cores, polypoid ureteritis/pyelitis, and reactive urothelium mimicked urothelial neoplasms. Smith et al.15 found that changes in the grade of UC between the diagnostic biopsy and resection specimen occurred in 24 of 65 biopsies (37%), including 9 in which diagnosis changed from low to high. Similarly, in our study, 31.4% of cases in which LGUC was initially diagnosed on biopsy were upgraded to HGUC on the nephroureterectomy specimen. Interestingly, 36.8% of HGUC from UT biopsy were downgraded to LGUC after nephroureterectomy. The discrepancies could be attributed to the limited size of the

Discussion In this study, we found that detecting HGUC in UT cytology is comparable to that of UT biopsies. Not surprisingly, sensitivity was higher for HGUC, highstage disease, and UT sampling due to the distinguishable morphology of high-grade tumor cells, tendency of exfoliation, and targeted sampling. Our study compares favorably with data on sensitivity of selective UT cytology for detecting UTUC reported in previous studies that range between 43% and 79%.3,10,16 Sedlock et al.17 stated the false-negative results could be as high as 50% for low-grade urothelial neoplasms. A contemporary multicenter study10 found that positive or atypical cytology obtained via selective ureteral catheterization yielded the highest diagnostic accuracy for high-grade or muscleinvasive UTUC (ie, 69% and 76% sensitivity for high-grade and muscle-invasive UTUC, respectively) with positive predictive values >85%. When considering only positive cytology, they noted a low overall sensitivity and positive predictive value for high-grade UTUC (56% and 54%, respectively) and muscle-invasive UTUC (62% and 44%, respectively). At our institution, the urologists

Table 3

Relationship between urinary cytology results and tumor size/staging (invasion).

Tumor grading Low-grade UTUC High-grade UTUC

Cytology results Positive Negative Positive Negative

Tumor size, cm 2.2 4.0 3.6 3.1

Abbreviation: UTUC, upper tract urothelial carcinoma.

   

1.5 2.6 2.1 1.6

Tumor staging, n (%) Ta/Tis

T1

22 13 7 5

6 11 13 5

(78.6) (54.2) (9.7) (33.3)

(21.4) (45.8) (18.1) (33.3)

T2

T3

T4

0 0 11 (15.2) 3 (20)

0 0 39 (54.2) 2 (13.4)

0 0 2 (2.8) 0

Comparative study of upper tract urinary cytology and surgical biopsy Table 4

7

Relationship between ureteroscopic biopsy results and tumor size/staging (invasion).

Tumor grading Low-grade UTUC High-grade UTUC

Biopsy results Positive Negative Positive Negative

Tumor size, cm 3.9 3.4 3.4 4.4

   

2.8 1.0 1.7 1.8

Tumor staging, n (%) Ta/Tis

T1

T2

T3

T4

7 3 2 1

7 1 5 1

0 0 4 (16.0) 1 (20.0)

0 0 13 (52.0) 2 (40.0)

0 0 1 (4.0) 0

(50.0) (75.0) (8.0) (20.0)

(50.0) (25.0) (20.0) (20.0)

Abbreviation: UTUC, upper tract urothelial carcinoma.

biopsy, absence of papillary fronds, crush artifact, and distorted architecture.4 The shortcomings of biopsy and the wider area sampled by cytology, make these 2 modalities complementary in detecting UTUC. Renshaw3 compared ureteral washing and biopsy among 39 patients in the community setting. The sensitivity was 79% for washings and 71% for biopsies, for a combined sensitivity of 100%. Some studies further suggest adding preoperative imaging results (local invasion and hydronephrosis) can improve the prediction of muscleinvasive and non-organ-confined disease in patients with UTUC.1,2 By studying 172 patients with advanced UTUC, Brien et al.2 found combining all 3 tests (imagining, biopsy, and cytology) incrementally improved the prediction of the stage of UTUC. Abnormality of all 3 tests had 89% and 73% positive predictive value for muscle-invasive and nonorgan-confined UTUC, respectively, but when all tests were normal, the negative predictive value was 100%. However, because we achieved 100% sensitivity by combining cytology and biopsy data, we could not evaluate any incremental increase in sensitivity obtained by adding imaging data. By evaluating 143 urinary specimens and 54 biopsies from UTUC patients in our institution, we found the sensitivity of selective UT cytology for HGUC is comparable to the endoscopic biopsy (80.6% versus 82.9%, P Z 0.07), and higher than that reported by Messer et al.10 (80.6% versus 71%). Although the calculation of cytology sensitivity only included positive results without combining the atypical cytology, AUC on UT cytology has high predictive value for developing UTUC according to 1 of our previous studies,18 in which we found the overall rate of atypia in urine specimens to be 8.1%. In this study, among initial AUC specimens, 21% of cases progressed to positive cytology or surgical pathology; however, UT specimens had the highest percentage (38%) of progression to HGUC. Therefore, we concluded that the diagnosis of atypia in this specimen group is associated with a higher rate of malignancy and should therefore be managed more aggressively. At our institution, the rate of malignancy associated with a cytologic diagnosis of “suspicious for urothelial carcinoma” (corresponding to AUC-H) 19-21 is very high, justifying their similar management to the “positive for urothelial carcinoma” diagnosis. Thus, we combined both groups into the positive category. Piaton et al.19 found that conservatively

treated patients with AUC-H more frequently developed high-grade lesions than did those with AUC-US ([atypical urothelial cells of undetermined significance], 54.1% versus 16.7%). Nuclear hyperchromasia, a nuclear-to-cytoplasmic ratio >0.7 and the combination of both were the more informative diagnostic criteria, all with P < 0.01. According to VandenBussche et al.,21 of specimens diagnosed with AUC-H, 95% ultimately were diagnosed with HGUC on follow-up biopsy. The terminology “AUC-H” as defined by these groups has essentially the same morphologic criteria for the category “suspicious for HGUC” used at our institution, which is the accepted terminology for the Paris System for Reporting Urinary Tract Cytology. We previously studied the role of cystoscopy and urine cytology in surveillance of LGUC’s recurrence.22 In this study, 76 patients were identified who had histologic follow-up material available, and 49% of those patients demonstrated progression or recurrence of UC. Cystoscopic findings, such as the presence of multiple lesions, together with concurrent positive or suspicious urine cytology, were associated with recurrence or progression of LGUC. The study findings may help to identify the patient’s high risk for development of HGUC. Tanaka et al.23 also investigated whether the status of preoperative urine cytology could provide additional prognostic information following radical nephroureterectomy. They included 474 patients with primary nonmetastatic UTUC (pT4a N0 M0) and a median follow-up period of 35 months. The results revealed that only the incidence of intravesical recurrence was significantly associated with the status of urine cytologic evaluation (P Z 0.024); the intravesical recurrence-free survival rates at 1 and 3 years following radical nephroureterectomy were 61.4% and 46.2% in patients with positive urine cytology and 71.1% and 51.6% in their counterparts, respectively. The study concluded that the prognostic value of positive urine cytology in patients with primary UTUC and preoperative positive urine cytology may be associated with a significant increase in the prevalence of intravesical recurrence following radical nephroureterectomy. However, no similar data from ureteroscopy has been reported to date. As expected, the cytology sensitivity for LGUC was not as promising as for HGUC. The low sensitivity of cytology for low-grade tumor is due to the similarity of cell morphology between low-grade tumor cells and benign urothelial cells. In addition, the higher sensitivity of UT

8 biopsy can also be explained by a selection bias as the UT biopsies are usually performed when suspicious lesions are visualized. By combining cytology and biopsy, the diagnostic sensitivities were increased to 87.5% for low-grade UTUC and 100% for high-grade UTUC. The results of our study are consistent with the findings reported by Renshaw.3 The consistency between biopsy and final surgical diagnosis on tumor grading were 63.2% for lowgrade UTUC and 68.6% for high-grade UTUC, respectively, which were lower than the results reported by Keeley et al.16 (90% and 91.6%, respectively). Tables 3 and 4 demonstrate that tumor grade corresponds well with tumor stage. A total of 54.2% of positive cytology in high-grade tumors has been staged as T3 in final pathology, whereas 78.6% of positive cytology in low-grade tumors has been staged as Ta/Tis. Because both cytology and biopsy showed higher sensitivities for high-grade UTUC, a positive cytology and high-grade tumor on biopsy may indicate the tumor has a higher stage. The negative cytology specimens in patients with high-grade UTUC were reviewed and no additional cases of AUC or above were detected. We therefore believe that the false negative results were due to sampling error. Several studies have shown that UroVysion/fluorescent in situ hybridization (FISH) study can improve the preoperative diagnosis of UTUC with sensitivity ranging between 52% and 78.9%.17,24,25 According to Xu et al.,25 the sensitivity of the combined cytology and FISH analysis reached 85.9%. Combined cytology with FISH analysis is a powerful tool and may improve the sensitivity and possibly the specificity of the UTUC diagnosis. Study results of Muus Ubago et al.26 showed the sensitivity of FISH among the “atypical urothelial cells present” specimens diagnosed by cytology is 54.8% with specificity of 84.7%. The study also showed the positive predictive value of FISH for “atypical urothelial cells present” specimens was 54.8%, compared with the negative predictive value of 90.3%. We concluded that FISH is not a replacement for cytologic analysis, but an adjunct to urine cytology.

Conclusions Both cytology and biopsy showed higher sensitivity in detecting high-grade UTUC than LGUC. The sensitivities of UT cytology and ureteroscopic biopsy in detecting highgrade UTUC were comparable. The sensitivity was greatly improved when these diagnostic modalities were combined. Compared with lower tract cytology sampling, the selective UT cytology evaluation had superior sensitivity in detecting UTUC. Approximately two-thirds of biopsy grading was consistent with grade on the final surgical diagnosis.

Funding sources No specific funding was disclosed.

L. Wang et al.

Conflict of interest disclosures The authors have nothing to disclose.

References 1. Favaretto RL, Shariat SF, Savage C, et al. Combining imaging and ureteroscopy variables in a preoperative multivariable model for prediction of muscle-invasive and non-organ confined disease in patients with upper tract urothelial carcinoma. BJU Int. 2012;109:77e82. 2. Brien JC, Shariat SF, Herman MP, et al. Preoperative hydronephrosis, ureteroscopic biopsy grade and urinary cytology can improve prediction of advanced upper tract urothelial carcinoma. J Urol. 2010;184: 69e73. 3. Renshaw AA. Comparison of ureteral washing and biopsy specimens in the community setting. Cancer. 2006;108:45e48. 4. Tavora F, Fajardo DA, Lee TK, et al. Small endoscopic biopsies of the ureter and renal pelvis: pathologic pitfalls. Am J Surg Pathol. 2009;33: 1540e1546. 5. Lotan Y, Roehrborn CG. Sensitivity and specificity of commonly available bladder tumor markers versus cytology: results of a comprehensive literature review and meta-analyses. Urology. 2003;61:109e118. 6. Brown FM. Urine cytology: is it still the gold standard for screening? Urol Clin North Am. 2000;27:25e37. 7. Ramakumar S, Bhuiyan J, Besse JA, et al. Comparison of screening methods in the detection of bladder cancer. J Urol. 1999;161:388e394. 8. Karakiewicz PI, Benayoun S, Zippe C, et al. Institutional variability in the accuracy of urinary cytology for predicting recurrence of transitional cell carcinoma of the bladder. BJU Int. 2006;97:997e1001. 9. Williams SK, Denton KJ, Minervini A, et al. Correlation of upper-tract cytology, retrograde pyelography, ureteroscopic appearance, and ureteroscopic biopsy with histologic examination of upper-tract transitional cell carcinoma. J Endourol. 2008;22:71e76. 10. Messer J, Shariat SF, Brien JC, et al. Urinary cytology has a poor performance for predicting invasive or high-grade upper-tract urothelial carcinoma. BJU Int. 2011;108:701e705. 11. Chen GL, El-Gabry EA, Bagley DH. Surveillance of upper urinary tract transitional cell carcinoma: the role of ureteroscopy, retrograde pyelography, cytology and urinalysis. J Urol. 2000;164:1901e1904. 12. Thompson RH, Krambeck AE, Lohse CM, Elliott DS, Patterson DE, Blute ML. Elective endoscopic management of transitional cell carcinoma first diagnosed in the upper urinary tract. BJU Int. 2008;102: 1107e1110. 13. Soderdahl DW, Fabrizio MD, Rahman NU, Jarrett TW, Bagley DH. Endoscopic treatment of upper tract transitional cell carcinoma. Urol Oncol. 2005;23:114e122. 14. Sedlock DJ, MacLennan GT. Urine cytology in the evaluation of upper tract urothelial lesions. J Urol. 2004;172:2406. 15. Smith AK, Stephenson AJ, Lane BR, et al. Inadequacy of biopsy for diagnosis of upper tract urothelial carcinoma: implications for conservative management. Urology. 2011;78:82e86. 16. Keeley FX, Kulp DA, Bibbo M, McCue PA, Bagley DH. Diagnostic accuracy of ureteroscopic biopsy in upper tract transitional cell carcinoma. J Urol. 1997;157:33e37. 17. Chen AA, Grasso M. Is there a role for FISH in the management and surveillance of patients with upper tract transitional-cell carcinoma? J Endourol. 2008;22:1371e1374. 18. Muus Ubago J, Mehta V, Wojcik EM, Barkan GA. Evaluation of atypical urine cytology progression to malignancy. Cancer Cytopathol. 2013;121:387e391. 19. Piaton E, Decaussin-Petrucci M, Mege-Lechevallier F, Advenier AS, Devonec M, Ruffion A. Diagnostic terminology for urinary cytology reports including the new subcategories “atypical urothelial cells of undetermined significance” (AUC-US) and “cannot exclude high grade” (AUC-H). Cytopathology. 2014;25:27e38.

Comparative study of upper tract urinary cytology and surgical biopsy 20. Rosenthal DL, Vandenbussche CJ, Burroughs FH, Sathiyamoorthy S, Guan H, Owens C. The Johns Hopkins Hospital template for urologic cytology samples: part Idcreating the template. Cancer Cytopathol. 2013;121:15e20. 21. VandenBussche CJ, Sathiyamoorthy S, Owens CL, Burroughs FH, Rosenthal DL, Guan H. The Johns Hopkins Hospital template for urologic cytology samples: parts II and III: improving the predictability of indeterminate results in urinary cytologic samples: an outcomes and cytomorphologic study. Cancer Cytopathol. 2013;121: 21e28. 22. Jackson J, Barkan GA, Kapur U, Wojcik EM. Cytologic and cystoscopic predictors of recurrence and progression in patients with lowgrade urothelial carcinoma. Cancer Cytopathol. 2013;121:398e402.

9

23. Tanaka N, Kikuchi E, Kanao K, et al. The predictive value of positive urine cytology for outcomes following radical nephroureterectomy in patients with primary upper tract urothelial carcinoma: a multi-institutional study. Urol Oncol. 2014;32:48.e19e48.e26. 24. Johannes JR, Nelson E, Bibbo M, Bagley DH. Voided urine fluorescence in situ hybridization testing for upper tract urothelial carcinoma surveillance. J Urol. 2010;184:879e882. 25. Xu C, Zeng Q, Hou J, et al. Utility of a modality combining FISH and cytology in upper tract urothelial carcinoma detection in voided urine samples of Chinese patients. Urology. 2011;77:636e641. 26. Muus Ubago J, Wojcik EM, Barkan GA. Correlation of fish with atypical urine cytology and its progression to malignancy. Mod Pathol. 2013;26(S2):105A.