Diagnostic status and new explorations in upper urinary tract urothelial carcinoma: a literature review.

Upper urinary tract urothelial carcinomas (UTUCs), including renal pyelocaliceal and ureteral cancers, are part of the broader category of urothelial cancers (UCs) alongside bladder cancers (BCs). In Western countries, UTUCs represent 5–10% of UCs ( 1 ), and in East Asia, their prevalence can reach up to 43% due to the use of aristolochic acid in herbal medicine ( 2 - 5 ). These carcinomas are characterized by their deep location and insidious onset, complicating early detection. About 50–60% of cases are muscularly invasive at diagnosis ( 6 ), and 20–30% present with lymph node metastasis ( 7 , 8 ). Extensive statistical analysis shows that advanced stages and lymph node involvement often correlate with a poor prognosis; 5-year cancer-specific survival rates are below 50% for pT2/pT3 patients and under 10% for pT4 cases ( 6 ). In clinical practice, current diagnostic standards primarily depend on urinary cytology, computed tomographic urography (CTU), and traditional white light ureteroscopy (WL-URS). However, these methods face significant challenges: urinary cytology suffers from low sensitivity; CTU struggles to effectively detect flat lesions and to distinguish between benign and malignant conditions; WL-URS, despite its diagnostic potential, is costly, technically challenging, and associated with patient discomfort as well as risks of surgical complications and tumor implantation in the bladder ( 9 - 11 ). These challenges underscore the necessity for more efficacious diagnostic techniques. Recent advancements have introduced a variety of new diagnostic tools and molecular markers, including innovative endoscopic and biopsy methods, radiomics, immunocytology, molecular diagnostics, and plasma protein classifiers. These innovations aim to rectify the deficiencies of traditional methods, offering a more thorough diagnostic framework. This review aims to offer a concise overview of these new diagnostic methods and markers, their effectiveness, and their potential integration into clinical practice. We present this article in accordance with the Narrative Review reporting checklist (available at https://tau.amegroups.com/article/view/10.21037/tau-24-326/rc ).

Two authors (Z.S. and X.D.) performed an exhaustive literature review using the PubMed, Embase, and Web of Science databases, employing a mix of subject terms and free words. The search strategy is elaborated in Appendix 1 (using PubMed as an example) and in Table 1 . All studies on the epidemiologic and diagnostic evaluation of UTUCs within the period from 2000 to August 2024 were included. Other literature was obtained from lists of references and similar literature. Exclusions were made for non-English papers, inaccessible full texts, and types such as commentaries, opinion pieces, editorials, letters, and case reports. Studies whose methods and results were judged unreliable by two corresponding authors (L.Y. and G.Z.) were also excluded. Figure 1 shows a flowchart summarizing the selection process. Given the heterogeneity of the findings, a narrative synthesis was conducted instead of a quantitative meta-analysis. UTUC, upper urinary tract urothelial carcinoma. Literature screening flowchart. UTUC, upper urinary tract urothelial carcinoma.

Through a comprehensive database search and literature tracking, we initially obtained 1,647 papers. After a systematic screening, over 100 papers were included in this study. Preliminary analysis revealed that while some studies continue to assess traditional diagnostic tools like urocytology, CTU, and WL-URS, there is a growing interest in exploring and developing new diagnostic techniques and molecular markers. These include radiomics, immunocytology, WL-URS, and various DNA-, RNA-, and protein-based markers. Urinary cytology is valued for its simplicity and non-invasive nature. It often indicates high-grade UTUCs when positive results are observed, excluding BCs and other urothelial carcinomas ( 12 ). However, its diagnostic sensitivity varies, with sensitivities reported between 11% and 71.1%, and specificities from 80% to 100% ( 9 , 13 - 30 ). Techniques such as selective collection of upper urinary tract (UUT) specimens via ureteral cannulae ( 31 , 32 ) or repeated saline flushes of the UUT mucosa (barbotage cytology) ( 33 ) have shown potential to enhance sensitivity, though these methods require further research for validation. Immunocytology is an advanced form of urinary cytology. Originating from a small 2001 study, it employed three monoclonal antibodies (19A211, M344, LDQ10) targeting two mucins and one carcinoembryonic antigen for UTUC diagnosis. This approach demonstrated a notable increase in sensitivity compared to traditional urinary cytology, particularly in detecting low-grade and early-stage tumors, and when combined, these methods further improved sensitivity ( 14 ). More recently, a novel urinary tumor cell (UTC) assay developed by Wang’s team utilized a nanostructured polystyrene substrate to capture over 90% of urinary exfoliated cells from morning urine samples. These cells were immunostained with CK20, CD45, and CD11b, and analyzed through a fully automated, high-resolution cytological imaging system. A sample was deemed positive if it contained one or more urinary cells ( 34 ). This test achieved an overall sensitivity of 85% and a specificity of 92.5% for UTC detection, outperforming urinary fluorescence in situ hybridization (FISH), which recorded sensitivities of 37.5% and 61.5% in low-grade and low-stage (≤ pT1) tumors, respectively ( 35 ). The test successfully identified all seven pTa cases, highlighting its potential for early UTUC diagnosis and suggesting promising prospects for widespread clinical application. CTU has replaced intravenous pyelography as the preferred imaging technique for suspected UTUCs, due to its superior ability to detect non-enhancing stones, blood clots, and severe obstructions ( 36 , 37 ). The introduction of multidetector CTU has further enhanced diagnostic capabilities, providing more detailed images with reduced ionizing radiation exposure and shorter acquisition times ( 38 ). Meta-analyses indicate that CTU maintains high specificity, exceeding 88%, with pooled sensitivities of 92% and 96%, and specificities of 99% and 95%, respectively ( 38 , 39 ). It also demonstrates robust T-staging accuracy, with understaging and overstaging rates reported at 8% and 11.2%, respectively ( 40 ). However, CTU faces challenges in effectively detecting carcinoma in situ (CIS) and superficial tumor extensions ( 41 ), with reports indicating that CIS and non-papillary tumors comprise up to 21.6–28.7% and 15.1–27.5% of UTUCs cases, respectively ( 42 , 43 ). Challenges also arise in distinguishing some benign conditions like chronic inflammation, ureteral endometriosis, and cystic ureteropelvic inflammation ( 36 , 37 ). MRU is considered an alternative for patients allergic to iodinated contrast agents or unable to undergo CTU. Despite this, the use of gadolinium-based contrast agents in MRU poses a risk of nephrogenic systemic fibrosis, especially in patients with renal impairment. Furthermore, MRU does not match CTU in terms of diagnostic and staging accuracy, with reported sensitivities generally below 76% ( 44 - 46 ). Radiomics analyzes multidimensional information reflecting tumor characteristics that are difficult to discern with the naked eye by utilizing high-throughput mining to diagnose the disease. Recently, three studies on UTUCs demonstrated that radiomics is effective for preoperative pathologic staging and grading, showing marked superiority to URS biopsy ( 47 - 49 ). Goto et al. analyzed CT texture in 86 patients, noting a significant difference in the histogram of CT attenuation numbers: muscle-invasive tumors exhibited a multi-peaked shape, whereas non-muscle invasive tumors displayed a single-peaked shape ( 47 ). Alqahtani et al. conducted a CTU-based radiomics analysis with 106 patients, achieving 84% and 83% sensitivity and 93% and 76% specificity, respectively, in predicting tumor grade (low-grade and high-grade) and staging (early stage, Ta-T1, and late stage T2-T4) ( 48 ). Similarly, Zheng’s study involving 140 patients used a radiomics-based machine learning model to predict tumor grading, achieving an area under the curve (AUC) of 0.914 and 0.903 in the training and validation sets, respectively ( 49 ). WL-URS allows direct visualization of the location, structure, and size of suspicious lesions, and facilitates biopsy, offering higher diagnostic accuracy and specificity compared to CTU ( 41 ). Despite its advantages, WL-URS has several limitations and risks, including: (I) a lower detection rate for CIS compared to CTU ( 41 , 43 , 50 ); (II) conventional ureteral forceps often fail to grasp subepithelial connective and deep muscular tissues, leading to inadequate staging and grading. A multicenter study reported only a 34.5% concordance rate between URS staging and grading and postoperative pathology results ( 51 ), with another meta-analysis highlighting high rates of understaging and undergrading at 46% and 32%, respectively ( 52 ); (III) URS is costly, technically demanding, and associated with significant risks, including infection, hemorrhage, and local injury. The literature indicates that the complication rate of URS is approximately 9–11% ( 53 ), with ureteral perforation accounting for 1–4% of cases, although most are localized ( 10 ); (IV) URS biopsies may cause tumor implantation in the bladder, with several studies indicating that URS biopsies significantly increase the rate of intravesical recurrence ( 11 , 54 - 57 ). NBI enhances the visualization of tumor boundaries and vascular structures by filtering broadband light to select narrow bands at 415 and 540 nm. This technique increases the detection rate of early lesions like pTa and CIS but may also result in a higher rate of false positives, leading to unnecessary biopsies ( 58 , 59 ). OCT provides imaging up to a depth of 2 mm, proving useful for preoperative staging and grading ( 60 ). A prospective study reported preoperative grading and staging accuracies of 83% and 88%, respectively, outperforming conventional biopsy ( 61 ). However, its limited depth restricts its applicability in advanced diseases. Originally employed in tracheal and intestinal examinations, CLE provides dynamic images of cellular structures and morphology using fluorescein staining ( 62 ). Its efficacy and safety in diagnosing UTUCs have been validated in studies with small sample sizes ( 62 , 63 ). Recent evaluations by Freund et al. reported CLE grading accuracies between 72% and 88% ( 64 ). However, CLE’s effectiveness relies heavily on image quality and necessitates specialized training. PDD utilizes 5-aminolevulinic acid (5-ALA), which is metabolized to protoporphyrin IX (PpIX) in vivo . Cancer cells deficient in ferrous chelatase accumulate PpIX, which fluoresces red under violet light ( 65 ). PDD significantly enhances diagnostic sensitivity, especially in flat lesions, CIS, and dysplasia. Nevertheless, it is susceptible to false positives due to local inflammation and experiences a considerable incidence of adverse effects, such as transient hypotension, photosensitized facial rash, and deranged liver enzymes, affecting up to 25% of patients ( 66 - 69 ). In clinical practice, devices for pelvic ureteral sampling include anterior and posterior biopsy forceps and tipless anterior and lateral grasping basket sets. Posterior biopsy forceps overcome the challenges of small-caliber anterior biopsy forceps, such as low tissue volume and high fragmentation, thus enhancing sample collection. The basket sets are particularly effective at grasping papillary tumors ( 70 , 71 ). Moreover, posterior large-caliber biopsy forceps and nested baskets secure higher quality tissue than anterior biopsy forceps, although nested baskets are somewhat less efficient and tend to produce larger artifacts ( 71 , 72 ). Klein et al. recently introduced a novel frozen biopsy technique that achieves the highest biopsy tissue volume and quality with the smallest area of artifacts in isolated renal pelvic ureteral tissues. This technique uses CO 2 to form ice crystals at the tip of a metal probe under a pressure drop, which then envelop the target tissue and adhere to the metal, facilitating tissue retrieval by suddenly withdrawing the prob ( 72 ). Additionally, Klett et al. developed a “formal processing” biopsy technique, which involves placing a cold cup clamp at the lesion’s base, closing it, and advancing it by 3–10 mm before withdrawal. This method has significantly improved tissue yields and diagnostic accuracy for lesions of 1 cm or larger ( 73 ). Recently, Fukumoto’s phosphorylated ribosomal protein S6-based 3D imaging and pathology analysis of URS biopsy tissues has enabled a more accurate predictive grading of tumors while avoiding cumbersome tissue sectioning, particularly beneficial in UTUCs, which are typically smaller and more friable ( 74 ). In UCs, urine directly contacts the tumor and is less complex than blood, facilitating UC diagnosis by detecting abnormalities such as tumor-derived DNA, mRNAs, and microRNAs (miRNAs). Research indicates that first-morning urine contains the most genetic information. The supernatant, comprising fragmented cell-free tumor nucleic acids and other derivatives, proves more effective than the sediment, which mainly holds detached cancer cells and their debris ( 75 , 76 ). Several studies have demonstrated that urine more accurately reflects the genetic abnormalities of tumor tissues in patients with UTUCs ( 3 , 42 , 77 , 78 ). Disparities in genetic abnormalities between urine and tumor samples may be due to inter-tumor and intra-tumor heterogeneity, clonal proliferation of the urinary epithelium, and insufficient aberrant cells or DNA in the urine ( 42 , 77 ). Significant advancements in urinary genetic abnormality tests have been made in diagnosing BCs, with tests such as Urovysion, EpiCheck test, and Xpert ® BC test receiving approval from the US Food and Drug Administration (FDA) and Conformitè Europëenne (CE) marking ( 79 ), advances in these tests for UTUCs are outlined below: FISH utilizes fluorescently labeled specific oligonucleotide fragments as probes to detect abnormalities in chromosomes 3, 7, 17, and 9p21 in urinary exfoliated cells, aiding in the diagnosis of urothelial carcinomas (UCs). FISH’s diagnostic value for UTUCs has been extensively explored, showing a wide range of sensitivities from 35% to 100%, generally outperforming urocytology, with specificities from 80% to 100% ( 9 , 15 , 18 - 20 , 22 - 29 , 80 , 81 ). However, most FISH studies involve small sample sizes and rarely evaluate low-grade and early-stage tumors ( 15 , 35 , 82 ). Noteworthy studies include the Sassa study and Aalami’s meta-analysis. The Sassa study, involving 75 patients with high-grade UTUCs, reported FISH sensitivities and specificities of 60% and 84%, respectively, compared to urocytology’s 28% sensitivity and 100% specificity ( 83 ). Aalami’s meta-analysis, including data from 1,067 patients, presented pooled sensitivity and specificity rates of 72% and 95%, respectively ( 84 ). Additionally, several studies have emphasized FISH’s role in predicting tumor pathological grading and intravesical recurrence ( 85 - 87 ). Springer et al. investigated aneuploidy by analyzing copy number changes on 39 chromosome arms in urine from 56 UTUCs patients. Alterations were found in 22 cases, with a detection sensitivity of 39%. The most frequently altered chromosome arms included 1q, 7q, 8q, 17p, and 18q ( 77 ). High-throughput sequencing of UTUCs tissues and urinary DNA has revealed significant mutations in genes such as FGFR3 , TERT , KMT2D , KDM6A , CDKN2A , TP53 , ARID1A , RAS , and CCND1 ( 42 , 77 , 78 , 88 ). FGFR3 , HRAS , and TERT mutations are more common in patients with non-muscle-invasive UTUCs, while TP53 and CCND1 mutations occur more frequently in muscle-invasive UTUCs. TERT and PIK3CA mutations are more prevalent in lower-grade tumors, while TP53 mutations are significantly higher in higher-grade tumors ( 42 , 78 ). TERT mutations in UTUCs urine primarily occur at two hotspots, 66 and 88 bp upstream of its transcription start site (g.1295228C>T and g.1295250C>T) ( 16 , 77 ). The frequency of mutations varies among different populations, with FGFR3 mutations being less frequent in the Chinese Han population ( 3 , 89 ). The distribution and frequency of these gene mutations assist in the diagnosis and staging of UTUCs. However, relying solely on single-gene or multi-gene testing does not fully address clinical needs. Recently, Fujii et al. achieved a sensitivity of 78% by sequencing urinary sediment-derived DNA in 43 UTUCs cases, significantly surpassing the 29.3% sensitivity of urinary cytology. This sensitivity remained consistent in 35 newly recruited patients, unaffected by previous UCs or the severity of pyuria and hematuria ( 42 ). When patients with severe urinary flow obstructions were excluded, sensitivity increased to 92.3%, with a specificity of 100% at a mutation detection rate of 74.5% ( 42 ). Urine-derived DNA methylation assays have been established for diagnosing BCs, such as the bladder EpiCheck assay, which utilizes 15 urinary DNA methylation biomarkers alongside a complex algorithm ( 79 , 90 , 91 ). However, these assays remain under-researched in UTUCs. To date, only the methylation of genes such as CDH1 , HSPA2 , RASSF1A , TMEFF2 , VIM , GDF15 , NRN1 , ONECUT2 , and bladder EpiCheck assays have been evaluated in UTUCs ( 78 , 92 - 97 ), with results detailed in Table 2 . Additionally, methylation in these genes is referenced against BCs, and extensive gene sequencing has shown significant differences in mutation frequency and burden between the two tumor types ( 42 , 88 ). Recently, Fujimoto et al. screened UTUCs tissues to identify 10 cytosine-phosphate-guanine (CPG) loci effective in differentiating BCs from normal uroepithelium. A diagnostic panel comprising these loci and neighboring sites achieved sensitivity, specificity, and AUC values of over 86.6%, 93.5%, and 0.959, respectively ( 99 ). The diagnostic efficacy was unaffected by clinicopathologic parameters such as gender, age, stage, grading, and lymph node metastasis, but it has not yet been validated in urine specimens. UTUCs, upper urinary tract urothelial carcinomas; PPV, positive predictive value; NPV, negative predictive value; AUC, area under the curve; LG, low grade; HG, high grade; UC, urothelial cancer; ddPCR, digital droplet polymerase chain reaction; NA, not assessed; PCR, polymerase chain reaction; Fast-SeqS, Fast-Sequencing System; NGS, next generation sequencing; NMIUC, non-muscle-invasive urothelial carcinoma; MIUC, muscle-invasive urothelial carcinoma; BS-RT-PCR, bisulfite-specific-RT-PCR; RT-PCR, real time-polymerase chain reaction; UUT, upper urinary tract; CIS, carcinoma in situ; qMSP, quantitative methylation-specific polymerase chain reaction; qPCR, quantitative polymerase chain reaction. From the limited study data available, the diagnostic efficacy of urine-derived DNA methylation testing for UTUCs is significantly better than that of urine genetic testing. Combining these two approaches can further enhance diagnostic accuracy. It is crucial to determine whether to proceed with CTU or URS based on methylation testing results, or to consider reducing the frequency of these procedures during follow-up. However, further evaluation is required. The Xpert ® BC assay is designed for the early diagnosis of BCs by qualitatively detecting the levels of five target mRNAs ( ABL1 , CRH , IGF2 , UPK1B , and ANXA10 ) in urine. The Xpert ® BC assay’s performance in diagnosing UTUCs was assessed in two similar studies featuring predominantly small samples of low-grade cases ( 97 , 98 ). In these studies, UUT urine was selected as the test specimen. The sensitivity reached 100% for both low-grade and high-grade patients, significantly higher than that of urocytology, FISH, and bladder EpiCheck tests, although the specificity was lower, at 4.5% and 16.7%, respectively. miRNAs are small non-coding, single-stranded RNA molecules. For example, serum miRNA371 is utilized in the diagnostic monitoring of testicular germ cell tumors and is included in various European clinical guidelines ( 100 ). In UTUCs, Tao et al. identified 13 significantly different miRNAs in 46 patients compared to 30 patients with hematuria. Ten of these miRNAs could differentiate between cancerous and hematuric patients with an AUC >0.8, with miR-664a-3p, miR-431-5p, and let-7c demonstrating the strongest performance ( 101 ). Kriebel et al. reported that miR-141 was overexpressed in both UTUCs serum and matched tissues ( 102 ). A larger study showed that miR-210 was effective in identifying oncogenic diseases in tissues ( 103 ). Recently, Urabe et al. identified 12 differential miRNAs in the serum of UC patients, with miR-4433a-3p and miR-6778-5p performing best in UTUCs ( 104 ). The diagnostic performance of these miRNAs for UTUCs is summarized in Table 3 . miRNAs, microRNAs; UTUCs, upper urinary tract urothelial carcinomas; PPV, positive predictive value; NPV, negative predictive value; AUC, area under the curve; PCR, polymerase chain reaction; NA, not assessed; qRT-PCR, quantitative reverse transcription polymerase chain reaction; LG, low grade; HG, high grade. In addition to their diagnostic applications, miRNAs also contribute to staging and grading. Tao et al. and Kriebel et al. observed that miR-664a-3p, miR-431-5p, miR-423-5p, miR-191-5p, miR-92a-3p, miR-16-5p, and let-7b-5p were significantly upregulated in patients with myxoid invasion, while miR-10a and miR-135 were downregulated ( 101 , 102 ). Browne et al. developed diagnostic models to predict high-grade (model ①) and myeloid invasion (model ②) in patients, by optimizing combinations of miRNAs identified in 158 UTUC tissues, achieving AUCs of 0.86 and 0.90, respectively ( Table 3 ) ( 105 ). Protein-based tumor markers are the most traditional and widely used diagnostic tools because they are easily accessible and minimally invasive. In UCs, the direct contact between the tumor and urine makes urine protein detection in urine a viable method. Recently, with advances in high-throughput sequencing technology, several UTUC-associated proteins and post-translational modifications have been detected and evaluated. However, these findings are mostly reported by single institutions. Below are some of the key proteins and modifications identified: SPP1 is part of the small integrin-binding ligand N-linked glycoprotein family. Li et al. discovered that SPP1 in UTUC tissues was the most significantly upregulated protein-coding gene (over 70-fold) among 948 differentially regulated genes compared to matched normal urinary tract epithelium ( 106 ). Enzyme-Linked Immunosorbent Assay (ELISA) tests on larger serum samples demonstrated that SPP1 was significantly correlated with pathological grade and T-stage, achieving an AUC of 0.838 for distinguishing cancerous from noncancerous cases, with sensitivities of 87% for muscle-invasive and 77.8% for high-grade disease ( 98 ). N-glycan modifications are common post-translational modifications. Kodama et al. used fast capillary electrophoresis with light-emitting diode-induced fluorescence to identify 26 N-glycan signals on serum immunoglobulins. UTUC scores constructed using these signals outperformed urocytology in UTUC detection, with sensitivity, positive predictive value (PPV), negative predictive value (NPV), and AUC at 90% specificity of 67.6%, 86.8%, 76.3%, and 0.868, respectively ( 107 ). Recently, Qu et al. conducted high-throughput measurements of plasma proteins from 362 UTUC patients and 239 healthy individuals, developing a 10-protein classifier model ( ANXA6 , CTSB , CD44 , SAA1 , HBZ , S100A8 , DPH5 , PAM , RAP1A , and THUMPD1 ) that effectively differentiated cancerous from non-cancerous cases, with sensitivity, specificity, and AUC of 99.7%, 99.8%, and 0.942, respectively ( 101 ). This model also performed well in an independent cohort with an AUC of 0.925 ( 108 ). Additionally, various 12-protein classifiers were constructed to predict the presence of muscle-invasive disease, all achieving AUCs greater than 0.73. The BTA test, an economical and rapid assay, detects human complement factor H-associated proteins in urine and is available in both qualitative and quantitative formats. Walsh et al. reported sensitivities and specificities of 82% and 89%, respectively, using the qualitative assay, significantly outperforming urinary cytology ( 13 ). A recent study of a Chinese population indicated sensitivities of 98% and 93%, and specificities of 63% and 81% for qualitative and quantitative tests, respectively, respectively ( 109 ). The NMP22 assay, which measures the concentration of NMP22 significantly higher in UC cells than in normal cells, showed sensitivities between 70–73.2% and specificities from 43.2–92%. Notably, this assay’s performance was consistent across samples from the UUT or bladder, although results may be influenced by hematuria, urinary tract infections, and instrumentation ( 17 , 22 , 109 ). The urinary p16/Ki-67 double immunolabeling technique uses biomarkers p16 and Ki-67 for diagnostic purposes. p16, a product of the CDKN2a gene on chromosome 9p21, regulates the cell cycle, while Ki-67 is a marker of cell proliferation. Sun et al. reported that the diagnostic sensitivity of simultaneous positivity of p16 and Ki-67 in UTUC was 53.1% (for low-grade and high-grade tumors 12.5% and 66.7%, respectively) and specificity was 100% ( 15 ).

Given the high incidence of certain diseases in East Asian countries and their characteristics—insidious onset, difficult diagnosis, easy invasion and metastasis, and poor prognosis—it is highly valuable to search for and develop safe, effective, and simple diagnostic tools. Historically, urine cytology, CTU/MRU and WL-URS have been common in clinical settings, yet they exhibit numerous drawbacks: urine cytology has low sensitivity; CTU can miss superficial extension diseases and carries the potential risk of contrast agents; and URS is costly, involves complications related to invasive procedures, and poses a risk of tumor implantation in the bladder. Fortunately, new techniques have partially addressed these shortcomings. For example, UTC tests have significantly enhanced diagnostic sensitivity for early-stage tumors by enriching urinary exfoliated cells with special materials; various new light source endoscopes have improved tumor visualization and detection, especially for CIS; and advancements in biopsy techniques and devices have enhanced the quality of biopsy specimens and preoperative staging and grading. Additionally, the performance of plasma protein classifiers has been notably impressive. However, we must acknowledge that new technologies also have drawbacks, such as: (I) many new techniques and markers have only been reported by single institutions or shown differentiation between cancerous cases and controls under experimental conditions. (II) Certain methods, such as the NBI test and urine mRNA tests, exhibit a high rate of false positives, which could lead to unnecessary further investigations or surgeries. (III) The efficacy of urinary molecular markers may be affected by hydration status, renal pathology, and medication use. (IV) Many molecular marker tests require sophisticated expertise and advanced laboratory facilities, demanding substantial financial resources.

In this review, we have summarized the benefits, risks, and limitations of UTUC-related diagnostic tools ( Table 4 ). Looking forward, our expectations for future research and the application of new technologies, particularly in molecular diagnostics, include: (I) expanded research and validation across multiple institutions and diverse populations, with performance evaluations in actual clinical settings. (II) Focused evaluations on patients with low-grade, low-stage, CIS, postoperative recurrences, and those with concurrent BCs. (III) Prospective multicenter comparative studies conducted alongside traditional methods such as CTU and WL-URS. (IV) Economic analyses via model construction to assess cost-effectiveness. (V) The development of high-throughput assays and analyses of genetic, protein, and epigenetic modifications in UTUCs to create diagnostic panels that are better tailored to the unique needs and characteristics of UTUCs. UTUC, upper urinary tract urothelial carcinoma; CTU, computed tomography urography; MRU, magnetic resonance urography; URS, ureteroscopy; WL-URS, white light ureteroscopy; CIS, carcinoma in situ; NBI, narrow-band imaging; OCT, optical coherence tomography; CLE, confocal laser endomicroscopy; PDD, photodynamic diagnosis; FISH, fluorescence in situ hybridization; BCs, bladder cancers; miRNAs, microRNAs; SPP1, plasma secreted phosphoprotein 1; NMP22, nuclear matrix protein 22; BTA, bladder tumor antigen.

The article’s supplementary files as