High-Grade Ureteroscopic Biopsy Is Associated with Advanced Pathology of Upper-Tract Urothelial Carcinoma Tumors at Definitive Surgical Resection

Abstract

Background and Purpose:

Accurate assessment of upper-tract urothelial carcinoma (UTUC) pathology may guide use of endoscopic vs extirpative therapy. We present a multi-institutional cohort of patients with UTUC who underwent surgical resection to characterize the association of ureteroscopic (URS) biopsy features with final pathology results.

Patients and Methods:

URS biopsy data were available in 238 patients who underwent surgical resection of UTUC. Biopsies were performed using a brush biopsy kit, mechanical biopsy device, or basket. Stage was classified as a positive brush, nonmuscle-invasive (<pT₂), or muscle invasive (MI; ≥pT₂). Grade was classified as low or high.

Results:

On URS biopsy, 88/238 (37%) patients had a positive brush, 140 (59%) had a diagnosis of non-MI, and 10 (4%) had MI disease. Biopsy results showed low-grade cancer in 140 (59%) and high-grade cancer in 98 (41%). Pathologic evaluation at surgical resection demonstrated non-MI tumors in 140 (59%) patients, MI in 98 (41%), and high-grade disease in 150 (63%). On univariate analysis, high URS biopsy grade was associated with high-grade (positive predictive value [PPV] 92%, P<0.0001) and MI (PPV 60%, P<0.0001) UTUC at surgery. URS biopsy stage, however, was associated with surgical pathology grade (P=0.005), but not MI (P=0.16) disease. On multivariate analysis, high URS grade, but not biopsy stage, was associated with high final pathology grade (hazard ratio [HR] 16.6, 95% confidence interval [CI] 7.0–39.5, P<0.0001) and MI UTUC (HR 3.6, 95% CI 2.1–6.8, P<0.0001).

Conclusion:

High URS biopsy grade, but not stage, is associated with adverse tumor pathology. This information may play a valuable role for risk stratification and in the appropriate selection of endoscopic management vs surgical extirpation for UTUC.

Introduction

U pper-tract urothelial carcinoma (UTUC), comprising cancers of the ureter and renal pelvis, accounts for 5% of all urothelial malignancies and 10% of renal tumors.¹ The gold standard management of UTUC lesions is radical nephroureterectomy (RNU) with resection of an ipsilateral bladder cuff. Oncologic outcomes with extirpative surgery demonstrate durable responses.^2,3

Endoscopic treatment of UTUC tumors had historically been reserved for patients with imperative indications for renal preservation. Contemporary data, however, suggest that elective consideration is reasonable in appropriately selected patients.^4

–8 In particular, both Soderdahl and colleagues⁶ and Sowter and associates⁷ highlighted that low-grade, nonmuscle-invasive (non-MI) UTUC tumors respond favorably to endoscopic therapy, while higher grade and stage lesions typically fail conservative treatments, necessitating RNU. Therefore, the ability to accurately predict pathologic stage and grade is paramount in the appropriate selection of primary therapy (RNU vs endoscopic), thereby obviating the sequelae of overtreatment (ie, renal unit loss) or the need for salvage therapy because of undertreatment in patients at high risk of endoscopic management failure.

Advances in ureteroscopic (URS) techniques have improved biopsy accuracy of upper-tract urothelial lesions. Nonetheless, obtaining a sufficient specimen for adequate interpretation continues to be a challenge. Renshaw⁹ highlighted this point by noting only a 71% sensitivity for URS biopsies to identify malignancy in 39 patients undergoing ureteral biopsy. Such issues have further generated conflicting reports regarding the ability for URS biopsy to accurately predict grade and stage of UTUC lesions.^10
–12 Many of these studies, however, are limited by relatively small sizes and are composed of data from biopsies performed by a single technique and interpreted by a single pathologist. Such results may not be generalizable to “real world” clinical practice. Therefore, in this large, multi-institutional cohort of patients undergoing extirpative surgery for UTUC, we sought to better determine the relationship between preoperative URS biopsy data and final pathologic disease characteristics.

Patients and Methods

Study population

Institutional Review Board approval was obtained at all participating institutions. The charts of 469 patients with clinically localized UTUC who underwent RNU or distal ureterectomy at five United States medical centers between September 1997 and August 2008 were reviewed retrospectively. RNU was performed by an open or minimally invasive approach, with management of the distal ureter per the surgeon's discretion. All distal ureterectomies were performed via an open approach. Patients who received neoadjuvant chemotherapy were excluded to minimize the potentially confounding impact on pathologic data. Data from URS biopsies were available in 238 of these patients, who form our cohort for analysis here.

URS biopsies

URS biopsies were performed using a: (1) brush biopsy kit (nondeflecting brush or metal deflecting brush, Boston Scientific, Natick, MA); (2) 3F mechanical biopsy device (Piranha, Boston Scientific, Natick MA); or (3) nitinol deflectable tip basket. URS biopsy stage was classified as: Positive brush cytology, non-MI (<T₂), or muscle invasive (MI; ≥pT₂). URS grade was classified according to the 2004 World Health Organization (WHO) criteria as low or high grade. The specific biopsy technique as well as number of biopsies performed was at the treating urologist's discretion.

Final pathology specimens

All specimens were histologically confirmed to be urothelial carcinoma by institutional genitourinary pathologists. Tumor classification was assessed according to the 2002 American Joint Committee on Cancer – Union International Contre le Cancer Tumor-Node-Metastasis classification. Tumor grading was assessed according to the 2004 WHO/International Society of Urologic Pathology consensus classification.

Statistical analysis

The chi-square test evaluated the association between URS biopsy data and final UTUC pathology results. Logistic regression methodology was used for prediction of MI and high-grade UTUC disease. Sensitivity, specificity, and predictive values were calculated using URS biopsy as the diagnostic test and tumor grade (high vs low) or stage (MI vs non-MI) as disease outcomes. All reported P values are two-sided, and statistical significance was set at ≤0.05. Statistical analysis was performed with SPSS (version 13.0, Chicago, IL).

Results

Clinical and pathologic data

Included in the final evaluable cohort of 238 patients were 154 men and 84 women with a median age of 70 years (range 34 to 96 y). Approximately 90% of patients in our cohort were managed by RNU. Forty percent of patients had tumors in the renal pelvis, 29% in the ureter, 31% had multifocal locations, and 3% had unclassified locations. On URS biopsy, 88 (37%) patients had a positive brush cytology. Additional biopsy results reported staging as non-MI in 140 (59%), and MI in 10 (4%) cases. With regard to URS biopsy grade, 140 (59%) biopsies were reported as low grade, while 98 (41%) were diagnosed as high grade.

Final pathologic analysis of surgical specimens demonstrated non-MI tumors in 140 (59%) patients and MI UTUC cancers in 98 (41%). Meanwhile, high-grade disease was found in 150 (63%) of the final tumor specimens (Table 1).

Table 1.

Clinical and Pathologic Characteristics of 238 Patients with Ureteroscopic Biopsy Data Available Before Definitive Surgery for Upper-Tract Urothelial Carcinoma

	No. (%)
Patients	238
Age
Median (range)	70 (34–96)
Sex
Male	154 (65)
Female	84 (35)
Surgery
RNU	213 (89)
DU	25 (11)
Tumor location
Renal pelvis	93 (40)
Ureter	67 (29)
Multifocal	72 (31)
Unspecified	6 (3)
URS biopsy stage
Positive brush	88 (37)
Nonmuscle invasive	140 (59)
Muscle invasive	10 (4)
URS biopsy grade
Low	140 (59)
High	98 (41)
Pathologic T stage
T₀	7 (3)
T_a	79 (33)
T_is	11 (5)
T₁	43 (18)
T₂	27 (11)
T₃	64 (27)
T₄	7 (3)
Pathologic tumor grade
Low	88 (37)
High	150 (63)

RNU=radical nephroureterectomy; DU=distal ureterectomy; URS=ureteroscopic.

Predictive accuracy of high-grade or MI URS biopsies

On univariate analysis, high URS biopsy grade was associated with high grade on the surgical specimen (positive predictive value [PPV] 92%, P<0.0001) as well as finding MI UTUC (PPV 60%, P<0.0001). There was no association between high URS biopsy grade and other clinicopathologic variables including patient age (P=0.53), sex (P=0.61), type of surgery (RNU vs distal ureterectomy, P=0.09), or tumor location (ureter vs renal pelvis vs multifocal, P=0.17).

When considering URS biopsy stage, we observed an association with surgical pathologic grade (PPV for MI URS biopsy 100%; PPV for positive brush URS biopsy 73%, P=0.005). URS biopsy stage, however, was not associated with surgical pathologic stage (PPV for MI URS biopsy 70%, PPV for positive URS brush 41%, P=0.16). Additional variables including age (P=0.63), sex (P=0.76), type of surgery (P=0.27), and location of tumor (P=0.19), were likewise not associated with URS biopsy stage.

In a multivariate model that controlled for age, sex, type of surgery, and tumor location, high URS grade (but not stage) was associated with high-grade (hazard ratio [HR] 16.6, 95% confidence interval [CI] 7.0–39.5, P<0.0001) and MI UTUC (HR 3.6, 95% CI 2.1–6.8, P<0.0001) on the final surgical specimen (Table 2).

Table 2.

Multivariable Model for Prediction of High-Grade and Muscle-Invasive Upper-Tract Urothelial Carcinoma on Final Pathologic Determination

	Final RNU Pathology Results
	High-grade UTUC			Muscle-invasive UTUC
Variables	HR	95% CI	P value	HR	95% CI	P value
URS grade	16.6	7.0–39.4	<0.001	3.8	2.1–6.8	<0.001
URS stage	1.1	0.7–1.8	0.74	1.3	0.7–2.1	0.39
Sex	1.0	0.5–2.0	0.95	1.8	1.0–3.3	0.06
Age at surgery	1.0	1.0–1.0	0.62	1.0	1.0–1.1	0.85
Tumor location: Ureter vs RP	1.0	0.6–1.5	0.93	1.2	0.8–1.8	0.29
Type of surgery	0.9	0.8–1.3	0.45	1.1	1.0–1.8	0.37

UTUC=upper-tract urothelial carcinoma; HR=hazard ratio; CI=confidence interval; URS=ureteroscopic; RP=renal pelvis.

Predictive accuracy of low-grade URS biopsies

Low-grade URS biopsies were collected in 140 patients. Low URS biopsy grade had a sensitivity of 91%, specificity of 60%, PPV of 54%, and negative predictive value (NPV) of 92% for low-grade UTUC tumors at final pathology determination. When considering predictive accuracy for non-MI UTUC at definitive resection, low URS biopsy grade had a sensitivity of 71%, specificity of 59%, PPV of 71%, and NPV of 60%.

Discussion

Accurate staging of UTUC tumors before definitive surgical therapy may permit tailoring of treatment modalities for individual patients. While RNU continues to represent the standard of care for UTUC, endoscopic ablation via either an antegrade percutaneous or retrograde URS approach has an expanding role in the treatment of UTUC.^4

–8 Specifically, endoscopic ablation is a reasonable strategy for patients with imperative indications, including an anatomic or functional solitary kidney, bilateral UTUC, baseline renal insufficiency, or significant comorbidities precluding radical surgery. In addition, selected patients with low-grade and low-volume UTUC disease and a normal contralateral kidney may also be suitable candidates for endoscopic management.^4,13

Central to the success of endoscopic management for UTUC is appropriate selection of low-grade and stage lesions that are less likely to have conservative therapy failure. In that regard, current strategies for preoperative staging of UTUC lesions remain limited. Historically, URS biopsy samples (while theoretically ideal by obtaining tissue confirmation of the tumor) provided varying levels of accuracy because of small sample sizes and associated pathologic challenges when interpreting such specimens.⁹

In 1997, Keeley and coworkers¹⁰ published results of 51 cases of UTUC diagnosed ureteroscopically and undergoing definitive extirpative surgery. These authors reported a grade-for-grade concordance between URS biopsy and final pathology results of approximately 90% (90% for low-grade and 92% for high-grade lesions). Furthermore, they observed an association between URS grade and pathologic stage—in 87% of low-grade URS specimens demonstrating a final pathologic stage of T_a or T₁ and 67% of high-grade URS specimens demonstrating a final stage of T₂ or T₃. Keeley and associates¹⁰ concluded that URS biopsy provided accurate information regarding the grade and stage of UTUC tumors.

Guarnizo and colleagues¹¹ published similar results from 40 UTUC tumors in patients undergoing a URS guided multibiopsy approach using either a 3F cup forceps or an 11.5 F resectoscope. In this study, URS biopsy grade accurately predicted final pathologic grade in 78% of cases. Despite obtaining lamina-propria depth specimens in 27/40 (68%) cases, however, 45% of patients with URS biopsy T_a lesions were upstaged on final pathology results to pT₁ or greater.

Most recently, Smith and coworkers¹² highlighted potential limitations of URS biopsy for characterizing final UTUC pathology results. Among 65 biopsied UTUC lesions, 37% had a change in grade and 38% a change in stage on repeated pathologic evaluation. Furthermore, in this series, 43% of patients were reclassified from low-grade, noninvasive disease to high-grade or invasive disease. These authors concluded that more than one-third of UTUC tumors managed conservatively had a change in grade or stage on repeated biopsy or at the time of final extirpative surgery.

Our data from more than 230 patients represent the largest published series evaluating the diagnostic accuracy of URS biopsy. We observed that, on univariate analysis, high URS biopsy grade had a predictive accuracy of more than 90% for high grade and 60% for MI UTUC on final RNU pathology determination. In addition, on multivariate analysis, high URS biopsy grade was associated with a 16-fold greater likelihood of high grade and an almost 4-fold greater risk of MI UTUC at the time of definitive surgery. Therefore, our data underscore the association between higher grade tumors at URS biopsy and advanced pathologic characteristics of UTUC tumors.

The data also highlight the challenges previously documented in the published literature regarding URS biopsy stage and final pathologic stage. We noted that of the 140 patients who had a diagnosis of non-MI tumors on URS biopsy, 55 (39%) were upstaged to MI UTUC at RNU. Furthermore, of the 10 patients who demonstrated MI tumor on URS biopsy, only 6 were found to have MI UTUC on final tumor pathology results. Accordingly, URS stage was not associated with high-grade or MI disease at final pathology determination.

While high URS biopsy grade associated well with adverse UTUC pathology, we found limitations when considering the predictive accuracy of low-grade URS biopsies. In particular, for the 140 low-grade URS biopsies, 61 (44%) had high-grade disease and 40 (29%) had MI disease. Therefore, a low URS biopsy grade only had a PPV value of 54% and 71%, respectively, for low-grade and non-MI UTUC in the final surgical specimen. Such observations underscore that the URS biopsy technique can be improved on for further diagnostic optimization.

Other preoperative variables of interest similarly have shown potential limitations when considered in isolation. That is, Messer and associates¹⁴ found that urinary cytology failed in many cases to accurately predict MI (PPV 44%) or high-grade (PPV 54%) UTUC cancers. When evaluating the URS appearance of tumors, El-Hakim and colleagues¹⁵ noted that subjective impressions of tumor grades during a URS procedure were inaccurate in 30% of the cases.

Imaging modalities have historically reported good accuracy for detecting regional or distant UTUC disease, although determination of local stage has been problematic. Both Ng and colleagues¹⁶ (hydronephrosis on CT or MRI) and Matin and associates¹⁷ (high frequency ureteroscopic endoluminal ultrasonography) have implicated such radiographic techniques as potential staging avenues for further exploration. Both techniques, however, require validation in future studies. Indeed, several different groups have now published data analyzing the predictive accuracy of multiple variables to preoperatively improve prediction of UTUC stage.^18
–20 Of particular interest, Brien and colleagues¹⁹ highlighted that the presence of high URS biopsy grade, positive urinary cytology, and ipsilateral hydronephrosis had an 89% PPV for MI UTUC, while normality of all three variables yielded a NPV of 100% for muscle invasion.

We acknowledge several limitations in this study. Because of the multi-institutional nature of this study, the URS biopsy technique varied among urologists, and biopsy specimens were interpreted by several different genitourinary pathologists. As a consequence, this study lacked a centralized pathologic review of biopsy and final RNU specimens. While these are potential limitations, we believe this lack of standardization also makes our data more generalizable to urologists in different practice environments. We further acknowledge the absence of an endoscopic cohort to determine success or failure of this modality based on biopsy characteristics. Therefore, assumptions of success or failure of endoscopic ablation were purely based on published reports. Our principal goal was to objectively characterize the correlation of URS biopsy with tumor pathology at surgical resection rather than study the effectiveness of endoscopic ablation.

Conclusions

In a contemporary large multi-institutional cohort, high URS biopsy grade, but not stage, was associated with adverse tumor pathology. URS biopsies may play a valuable role for risk stratification regarding ablative vs extirpative therapies for UTUC. Prospective evaluation is requisite to validate these findings.

Footnotes

Disclosure Statement

No competing financial interests exist.

Abbreviations Used

References

Jemal

, Siegel

, Ward

et al. Cancer statistics, 2009. CA Cancer J Clin, 2009; 59:225–249.

Margulis

, Shariat

, Matin

et al. Outcomes of radical nephroureterectomy: A series from the Upper Tract Urothelial Carcinoma Collaboration. Cancer, 2009; 115:1224–1233.

Hall

, Womack

, Sagalowsky

et al. Prognostic factors, recurrence, and survival in transitional cell carcinoma of the upper urinary tract: A 30-year experience in 252 patients. Urology, 1998; 52:594–601.

Chen

, Bagley

. Ureteroscopic management of upper tract transitional cell carcinoma in patients with normal contralateral kidneys. J Urol, 2000; 164:1173–1176.

Daneshmand

, Quek

, Huffmann

. Endoscopic management of upper urinary tract transitional cell carcinoma: Long-term experience. Cancer, 2003; 98:55–60.

Soderdahl

, Fabrizio

, Rahman

et al. Endoscopic treatment of upper tract transitional cell carcinoma. Urol Oncol, 2005; 23:114–122.

Sowter

, Ilie

, Efthimiou

, Tolley

. Endourologic management of patients with upper-tract transitional-cell carcinoma: Long-term follow-up in a single center. J Endourol, 2007; 21:1005–1009.

Elliott

, Segura

, Lightner

et al. Is nephroureterectomy necessary in all cases of upper tract transitional cell carcinoma? Long term results of conservative endourologic management of upper tract transitional cell carcinoma in individuals with a normal contralateral kidney. Urology, 2001; 58:174–178.

Renshaw

. Comparison of ureteral washing and biopsy specimens in the community setting. Cancer, 2006; 108:45–48.

10.

Keeley

, Kulp

, Bibbo

et al. Diagnostic accuracy of ureteroscopic biopsy in upper tract transitional cell carcinoma. J Urol, 1997; 157:33–37.

11.

Guarnizo

, Pavlovich

, Seiba

et al. Ureteroscopic biopsy of upper tract urothelial carcinoma: Improved diagnostic accuracy and histopathological considerations using a multi-biopsy approach. J Urol, 2000; 163:52–55.

12.

Smith

, Lane

, Larson

et al. Inadequacy of biopsy for diagnosis of upper tract urothelial carcinoma: Implications for conservative management. Urology, 2011; 78:82–86.

13.

Thompson

, Krambeck

, Lohse

et al. Endoscopic management of upper tract transitional cell carcinoma in patients with normal contralateral kidneys. Urology, 2008; 71:713–717.

14.

Messer

, Shariat

, Brien

et al. Urinary cytology has a poor performance for predicting invasive or high-grade upper-tract urothelial carcinoma. BJU Int, 2011; 108:701–705.

15.

El-Hakim

, Weiss

, Lee

, Smith

. Correlation of ureteroscopic appearance with histologic grade of upper tract transitional cell carcinoma. Urology, 2004; 63:647–650.

16.

, Shariat

, Lucas

et al. Does the presence of hydronephrosis on preoperative axial CT imaging predict worse outcomes for patients undergoing nephroureterectomy for upper-tract urothelial carcinoma. Urol Oncol, 2011; 29:27–32.

17.

Matin

, Kamat

, Grossman

. High-frequency endoluminal ultrasonography as an aid to the staging of upper tract urothelial carcinoma: Imaging findings and pathologic correlation. J Ultrasound Med, 2010; 29:1277–1284.

18.

Williams

, Denton

, Minervini

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; 22:71–76.

19.

Brien

, Shariat

, Herman

et al. Preoperative hydronephrosis, ureteroscopic biopsy grade and urinary cytology can improve prediction of advanced upper tract urothelial carcinoma. J Urol, 2010; 184:69–73.

20.

Favaretto

, Shariat

, Savage

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, 2011E-pub ahead of print.