Introduction

Clinical staging of upper tract urothelial carcinoma (UTUC) is often inaccurate given the limitations of ureteroscopic biopsy. The ability to identify patients with invasive disease would help select those who may benefit from more aggressive treatments. We sought to describe the prevalence of and identify preoperative risk factors for muscle invasive (MI)-UTUC.

Methods

We performed a retrospective review using a combined institutional cohort of all patients who underwent definitive surgical management for UTUC from 2000-2016. Patients were only included in the analysis if they underwent preoperative ureteroscopic biopsy. We identified several preoperative tumor and patient characteristics, including the details of biopsy. Our primary outcome was MI-UTUC at final surgical pathology. We used logistic regression to identify predictors of MI-UTUC and all p-values <0.05 were considered statistically significant.

Results

We identified 259 patients meeting selection criteria with median age at biopsy of 72 years (IQR 64-79). On biopsy, 63% of tumors were high-grade (HG) and lamina propria (LP) invasion was seen in 24%. 88% of patients were treated with radical nephroureterectomy and 12% had a segmental ureterectomy. On final surgical pathology, 46% of tumors had MI-UTUC. On multivariate analysis, HG and LP invasion on biopsy were each independently associated with presence of MI-UTUC (table). The combined presence of both LP and HG on biopsy (n=61) had a positive predictive value (PPV) of 85% and a negative predictive value of 66% for MI-UTUC. While HG and LP invasion had a high specificity (94%), it was relatively insensitive (43%). We conducted a separate analysis to identify risk factors for absence of HG and LP invasion on biopsy for patients with MI-UTUC. On multivariate logistic regression, absence of HG and LP invasion for patients with MI-UTUC was associated with biopsy tissue samples ≤ 1 mm in length (OR 4.0, 95% CI 1.2-13.1; p=0.02).

Conclusions

Findings of HG disease and LP invasion on ureteroscopic biopsy are independent risk factors for MI-UTUC and combine for a high PPV. Additionally, invasive tumors may be missed when ureteroscopic biopsy only captures small tissue samples. There are likely additional radiographic or ureterscopic tumor characteristics that are associated with MI-UTUC that we were unable to quantify.

Funding

None