Upper Urinary Tract Urothelial Cell Carcinoma

5. DIAGNOSIS

5.1. Symptoms

The diagnosis of UTUC may be incidental or symptom related. The most common symptom is haematuria [11]. Flank pain, due to clot or tumour tissue obstruction, can occur in 20–32% of cases [11]. Pre-operative symptoms at diagnosis are associated with a worse prognosis [92]. Systemic symptoms (including anorexia, weight loss, malaise, fatigue, fever, night sweats, and cough) in patients with UTUC should prompt evaluation for metastases associated with a worse prognosis [11].

5.2. Imaging

5.2.1. Computed tomography

Computed tomography (CT) urography has the highest diagnostic accuracy of the available imaging techniques [93]. A meta-analysis of 13 studies comprising 1,233 patients revealed a pooled sensitivity of CT urography for UTUC of 92% (CI: 0.85–0.96) and a pooled specificity of 95% (CI: 0.88–0.98) [94].

Rapid acquisition of thin sections allows high-resolution isotropic images images of both upper urinary tracts that can be viewed in multiple planes to assist with diagnosis without loss of resolution. Epithelial “flat lesions” without mass effect or urothelial thickening are generally not visible with CT.

The presence of enlarged LNs on CT is highly predictive of metastases in UTUC [95,96]. The risk of thoracic metastases is extremely low in low-risk UTUC (see section 6 for UTUC risk classification variables).

5.2.2. Magnetic resonance urography

Magnetic resonance (MR) urography is indicated in patients who cannot undergo CT urography, usually when radiation or iodinated contrast media are contraindicated [97]. The sensitivity of MR urography is 75% after contrast injection for tumours < 2 cm [97]. Computed tomography urography is more sensitive and specific for the diagnosis and staging of UTUC compared to MR urography [98].

5.2.3. ¹⁸F-Fluorodeoxglucose positron emission tomography/computed tomography

A retrospective multicentre publication on the use of ¹⁸F-Fluorodeoxglucose positron emission tomography/ computed tomography (FDG-PET/CT) for the detection of nodal metastasis in 117 surgically-treated UTUC patients reported promising sensitivity and specificity of 82% and 84%, respectively. Suspicious LNs on FDG-PET/CT were associated with worse recurrence-free survival (RFS) [99]. These results warrant further validation and comparison with MR and CT. FDG-PET can also be used to assess (nodal and distant) metastases in patients unfit for iodinated contrast media due to renal impairment and/or allergy.

5.3. Cystoscopy

Urethrocystoscopy is an integral part of the UTUC work-up to rule out concomitant BC [8,20].

5.4. Cytology and urinary markers

Voided cytology may indicate high-grade UTUC when bladder cystoscopy is normal, and in the absence of CIS in the bladder and prostatic urethra [1,100]. Voided urine cytology is less sensitive for UTUC than selectively obtained cytology from the affected upper tract [101]. In a recent study, barbotage cytology detected up to 91% of cancers [102]. Barbotage cytology taken from the renal cavities and ureteral lumina is preferred before application of a contrast agent for retrograde ureteropyelography as it may cause deterioration of cytological specimens [100,102]. Retrograde ureteropyelography remains an option to detect UTUC [87,103,104]. The sensitivity of fluorescence in situ hybridisation (FISH) for molecular abnormalities characteristic of UTUC is approximately 72–84% [105,106]. In a systematic review, including 25 studies on cytology and urinary markers, cytology and FISH were most commonly used [107]. FISH had comparable specificity (80-100%) and a higher sensitivity (35-86%) compared to cytology (11-71%). However, considering the wide ranges in sensitivity and specificity for both cytology and FISH, the authors concluded that these tests were suboptimal to rule out UTUC. A prospective study in 79 patients with suspicion of UTUC using upper tract urine collected just before URS, reported sensitivities for Xpert Bladder, FISH, Bladder Epicheck and cytology of 100%, 87%, 64% and 42% , respectively. Specificities were 4%, 82%, 79% and 94%, respectively [108]. FISH, Bladder Epicheck and cytology could be helpful as an ancillary tool to detect UTUC; however, further confirmation in well-designed prospective comparative trials is needed.

5.5. Diagnostic ureteroscopy

Flexible ureteroscopy (URS) is used if it is necessary to confirm the diagnosis of UTUC by visualising the ureter, renal pelvis and collecting system and to perform a biopsy of suspicious lesions. It is also essential for meticulous tumour mapping before considering kidney-sparing options for UTUC. Presence, appearance, multifocality and size of the tumour can be estimated during URS. In addition, ureteroscopic biopsies can determine tumour grade in over 90% of cases with a low false-negative rate, regardless of sample size [109]. However, undergrading and understaging leading to inaccurate risk stratification occurs with ureteroscopic diagnostic biopsy compared to nephroureterectomy specimens [87,110,111].

Ureteroscopy also facilitates selective ureteral sampling for cytology [104,112,113]. Stage assessment using ureteroscopic biopsy can be inaccurate, hence, combining ureteroscopic biopsy grade, imaging findings, and urinary cytology may help in the decision-making process between radical nephroureterectomy (RNU) and kidney-sparing approach [113,114]. In a meta-analysis comparing URS vs. no URS prior to RNU, 8 out of 12 studies found an increased risk for intravesical recurrence in those undergoing URS [115]. Performing a biopsy at time of URS was also identified as a risk factor for intravesical recurrence [115]. A second systematic review of 16 studies showed that URS alone was not significantly related to intravesical recurrence; whereas, URS with a biopsy significantly increased the risk for subsequent intravesical recurrence albeit without an impact on extra urinary tract recurrences and overall survival [116].

Technical developments in flexible ureteroscopes and the use of novel imaging techniques may improve visualisation and diagnosis of flat lesions [117]. Narrow-band imaging is a promising technique, but results are preliminary [118]. Optical coherence tomography and confocal laser endomicroscopy (Cellvizio^®) have been used in vivo to evaluate tumour grade and/or for staging purposes, with a promising correlation with definitive histology in high-grade UTUC [119,120].

5.6. Molecular Testing

FGFR 2/3 alterations should be tested for by NGS (see section 7.3.2.2.3) in the metastatic setting preferably from an invasive part of the tumour or metastatic site [121,122].

5.7. Summary of evidence and recommendations for the diagnosis of UTUC