Bladder cancer is the tenth most common malignancy worldwide, with an estimated 550,000–570,000 new cases and more than 200,000 deaths each year.1,2 It represents a major health and economic burden due to its high incidence, lifelong surveillance needs, and substantial risk of recurrence and progression. Urothelial carcinoma (transitional cell carcinoma) is by far the most common type of bladder cancer, accounting for approximately 90 % of primary bladder malignancies. These cancers often arise from the urothelial lining of the bladder and can present as non-muscle-invasive papillary tumors, carcinoma in situ, or muscle-invasive disease.3 Less common histologic subtypes include squamous cell carcinoma, adenocarcinoma, and small cell/neuroendocrine carcinoma, which are highly aggressive, often presenting at an advanced stage,4 as well as primary sarcomas or other mixed histologies. In addition, the urinary bladder may be secondarily involved by direct extension or metastatic spread from other primary tumors (most commonly gynecologic, colorectal, or prostate cancers), and these lesions can mimic primary bladder cancers both clinically and radiologically.5

Approximately 70 % of patients present with non-muscle-invasive bladder cancer, which generally carries a favorable prognosis but can have high rates of recurrence requiring repeated cystoscopies and adjunctive imaging staging.3,6 In contrast, approx. 30 % of cases present as muscle-invasive bladder cancer carrying a much poorer prognosis that requires aggressive management, including radical cystectomy, systemic chemotherapy, radiotherapy, and increasingly, immunotherapy or targeted agents.7 Accurate staging and restaging are therefore critical to guide treatment selection and to stratify patients appropriately for curative or palliative approaches.

Bladder cancer imaging is undergoing rapid evolution, with advances in both anatomical and molecular imaging techniques reshaping diagnosis, staging, treatment planning, and surveillance. Conventional anatomical imaging, such as cystoscopy, ultrasound, CT, and MRI remains central for assessing tumor location, depth of invasion, lymph node status, and distant metastases. Multiparametric MRI (mpMRI) of the bladder is aided by standardized scoring systems (e.g. VI-RADS) to improve local staging accuracy. However, these modalities can be limited in identifying flat lesions, active microscopic nodal disease, and in distinguishing treatment-related changes from cancer. This has driven interest in molecular imaging, particularly PET/CT and PET/MRI using FDG and other novel bladder-specific or immuno-PET agents, offering complementary functional and metabolic information while enabling better detection of occult nodal or distant metastases, treatment response assessment, and patient prognosis.8 Despite these advances, the application of PET to bladder cancer presents unique challenges: urinary excretion of many PET tracers hampers evaluation of the bladder wall, availability and cost limit widespread adoption of newer PET agents, standardized protocols and interpretation criteria are still evolving, and robust evidence supporting the integration of imaging biomarkers into clinical decision pathways remains incomplete. Ongoing research aims to refine techniques, develop novel tracers, and integrate imaging with genomic profiling data to achieve more precise and personalized management of patients with bladder cancer.

FDG PET/CT is increasingly being used in bladder cancer, particularly for cancer staging and restaging however it is also subject to clear limitations.

While FDG PET/CT has limited ability to localize primary tumors because of its intense competing urinary track excretion, FDG PET/CT improves detection of nodal and distant metastases compared with conventional CT in many patients. Thus, FDG PET can refine initial staging, accurately assess treatment response and detect occult recurrence, which can change patient management (e.g., systemic therapy vs local treatment). Technical advances, such as forced diuresis, bladder irrigation, dynamic or delayed PET imaging, or PET/MR fusion can help mitigate urinary tracer activity and improve visualization of the pelvis. Quantitative metrics on PET such as SUV-based parameters and metabolic tumor volume, and emerging radiomic approaches show promise for prognostication and treatment stratification. Key challenges still include the intense urinary excretion of FDG that may obscure the primary tumor and nearby lymph nodes, lack of fully standardized imaging protocols, and limited large prospective, disease-specific trials compared with other malignancies. Moreover, inflammatory or post-treatment changes can confound interpretation, and current guidelines still vary in recommending FDG PET/CT for routine use, so its use remains somewhat controversial and less firmly established than with other malignancies.

Cystoscopy with transurethral resection of bladder tumor (TURBT) remains the standard of care to evaluate a primary bladder tumor.9 Superficial lesions are directly visible with cystoscopy, and deep biopsies into muscularis propria are taken to determine the depth of invasion. However, the accuracy of TURBT-based staging is reduced by upstaging from non-muscle-invasive to muscle-invasive bladder cancer occurring in approximately 32 % of patients when a repeat cystoscopy with transurethral resection is performed.10 FDG PET has a limited role in evaluating the primary tumor mainly because of intense physiological excretion of FDG into the urine, which can obscure intravesical lesions, reducing the sensitivity for detecting the primary tumor and superficial wall involvement (Fig. 1). In a meta-analysis regarding the performance of FDG-PET/CT for detecting bladder primary lesions, Wang et al. reported a sensitivity of 80.0 % (95 % CI: 71.0 vs. 87.0 %) and a specificity of 84.0 % (95 % CI: 69.0 vs. 93.0 %). Various methodologies, such as forced diuresis or furosemide, bladder catheterization with continuous drainage, and delayed imaging have been investigated to improve lesion conspicuity but are problematic from a workflow perspective.11 Optimized PET protocols incorporating dual-phase imaging after forced diuresis and/or oral (500–1000 mL) or intravenous hydration (500 mL) after FDG injection11, 12, 13, 14, 15 have demonstrated lowered tracer activity in the urinary bladder that resulted in enhanced tumor visualization; for instance Ibrahim et al. studied 39 bladder cancer patients using delayed FDG PET/CT imaging after forced diuretics and observed residual bladder lesions in 12.8 % of early images and 58.9 % of the delayed images12; Higashiyama et al. produced similar results evaluating 25 bladder cancer patients with delayed FDG PET/CT after oral hydration and voiding-refilling and reported sensitivities of 24.0 % on early imaging and 92.0 % on delayed imaging for detecting residual invasive bladder cancer.13 Other authors, such as Yoon et al. evaluated the role of early dynamic FDG PET achieving sensitivities of 85 % for early dynamic PET (at 10 min after injection), 58 % for whole-body (at 60 min), and 61 % for delayed PET (at 120 min) in 52 patients with bladder cancer.16 However, none of these methods have led to the routine incorporation of FDG PET/CT into the standard staging algorithms for local T-staging.

Pelvic MRI is the preferred imaging modality for local T staging of bladder cancer owing to its superior soft-tissue contrast resolution and multiplanar capability compared to CT, enabling detailed evaluation of the bladder wall layers and peri‑vesical structures.17,18 High-resolution T2-weighted imaging, typically combined with diffusion-weighted imaging (DWI) and dynamic contrast-enhanced (DCE) sequences within a multiparametric MRI (mpMRI) protocol, enhances discrimination between non-muscle-invasive (≤T1) and muscle-invasive (≥T2) disease and facilitates assessment of extravesical extension (T3) and invasion of adjacent organs (T4).17 A systematic meta-analysis of 20 studies, including 1724 patients reported a pooled sensitivity of 0.92 (95 % confidence interval [CI] 0.88–0.95) and pooled specificity of 0.88 (95 % CI 0.78–0.94), respectively, for differentiating ≤T1 from ≥T2 disease.19 MRI is particularly valuable for differentiating viable tumor from post-biopsy inflammation or fibrosis and for delineating involvement of ureters, prostate, vagina, or pelvic sidewall. Limitations include reduced accuracy for very small or flat lesions (e.g., carcinoma in situ), susceptibility to motion artifacts, and diminished performance following intravesical therapy or extensive TURBT. Despite these challenges, pelvic MRI currently provides the most accurate noninvasive method for local bladder cancer staging and is increasingly integrated into preoperative planning and risk stratification algorithms.20

At the time of initial diagnosis, approximately 70 %−75 % of bladder cancers are non-muscle-invasive, while 20 %–25 % are muscle-invasive, and 5 %−10 % present with metastatic disease.21 Bladder cancer most commonly metastasizes via lymphatic spread, often in the pelvic lymph nodes, most commonly the obturator (74 %) and external iliac (65 %) nodes. Less common nodal involvement include pre-sacral (25 %), common iliac (20 %) and para-vesical (16 %) nodes.21 The 5-year survival for non-metastatic muscle-invasive bladder cancer is approximately 70 %, dropping to ∼6-9 % for distant metastatic disease.4,22 Thus, the pattern and burden of metastatic disease have major implications for prognosis, staging, and treatment selection.

Conventional anatomical imaging, primarily cystoscopy, CT and MRI remain the standard of care for evaluating local extension and lymph node status. For nodal staging, both CT and MRI rely predominantly on morphologic criteria (i.e. short-axis diameter typically ≥8–10 mm, irregular nodal contours, loss of the fatty hilum); as a result, small metastatic deposits in non-enlarged lymph nodes are missed, leading to moderate to low sensitivity despite relatively high specificity. Across studies, CT typically shows sensitivity of approximately 40-60 % and specificity of 80-95 % for detecting nodal involvement based on the dimensions of the nodes, with overall accuracy often in the 70-85 % range.23 MRI offers comparable or slightly improved performance, with reported sensitivity of 50-80 %, specificity 85-95 %, and overall accuracy often ranging from roughly 73 % to 92 %,24 the results reflecting similar reliance on size-based criteria. Collectively, these data reflect reliable detection of bulky metastatic nodes, but a substantial limitation in identifying micro-metastases within normal-sized nodes. This diagnostic gap has prompted increasing interest in metabolic imaging using FDG PET (Fig. 2). Meta-analysis data across 14 studies (six prospective, eight retrospective) encompassing 785 patients in newly diagnosed bladder cancer reported pooled sensitivity and specificity values of 0.57 and 0.92, respectively, for preoperative lymph node (LN) staging using FDG PET/CT.25 Compared with CT alone, FDG PET/CT offers modestly improved sensitivity with similar specificity for detecting nodal disease in bladder cancer. Prospective data support this pattern with FDG PET/CT demonstrating higher sensitivity than CT for pelvic nodal staging (e.g., 52 % vs 38 %) while maintaining nearly identical specificity (approximately ≥90 % in both modalities) .26,27 In a large cohort of 300 patients reported by Moussa et al. FDG PET showed substantially higher sensitivity than CT (40.3 % vs 13.4 %), albeit with slightly lower specificity (79.5 % vs 86.7 %), indicating a high rate of false positives and the potential need for confirmatory biopsy.28 In a retrospective study of 75 bladder cancer patients, FDG PET demonstrated slightly higher sensitivity than CT (60 % vs 46 %); however, CT achieved 100 % specificity, as no enlarged (≥10 mm) lymph nodes on CT were associated with negative histopathology. In contrast, FDG PET/CT specificity was 83 %, suggesting that confirmatory biopsy should still be considered as an adjunct to a positive FDG PET finding to avoid false positive diagnoses29 Importantly, FDG PET/CT can reveal occult metastatic nodal disease not visualized on CT, altering management in a subset of patients (avoiding cystectomy due to unsuspected metastases).29 Nonetheless, its overall sensitivity remains insufficient to reliably exclude microscopic nodal metastases. Notably, few studies have shown a clear advantage of FDG-PET/CT over CT along for LN detection, reporting similar sensitivity and specificity between the two modalities.30,31

There is growing interest in investigating the role of PET/MRI for assessing nodal involvement. PET/MRI is an emerging alternative to FDG PET/CT for lymph node staging in bladder cancer, but current evidence suggests it is at best comparable, not clearly superior for nodal assessment. In a prospective pilot of 24 bladder cancer patients, Rosenkrantz et al. highlighted the potential of FDG PET/MRI over MRI alone, to detect pelvic LN involvement with an accuracy of 95 % for PET/MRI versus 76 % for MRI alone, providing more accurate staging mainly in the case of equivocal findings on MRI alone.32 Eulitt et al. in a pilot study of 21 patients, found FDG PET/MRI may improve the diagnostic accuracy for the staging of bladder cancer.33 However, limitations of PET/MRI include higher cost, longer scan times, more limited availability, and greater technical complexity.

When bladder cancer metastasizes beyond the pelvis, the lungs are the most common site of distant spread, followed by the liver and skeletal system. Less frequently, metastatic disease involves the adrenal glands, peritoneum, or brain. Metastases to the bone can exhibit various patterns, including lytic, sclerotic, or mixed appearances (Fig. 3), and rarely, they can lead to spinal cord compression. Lung metastases typically present as multiple nodules, while less commonly, they may manifest as lung consolidations or lymphangitic carcinomatosis, characterized by nodular or irregular thickening of interlobular septa. Liver metastases commonly appear as multiple hypodense lesions on CT scans. Metastatic peritoneal disease may manifest as peritoneal nodules, thickening of peritoneal folds, peritoneal stranding, serosal metastasis, or ascites. In some instances, extensive serosal disease can lead to bowel obstruction.34 Overall, distant metastases occur in approximately 10–29 % of patients with bladder cancer and are associated with a poor long-term prognosis, with a reported 5-year survival rate of only 9 %.4,22

It has been suggested that disease progression occurs primarily through lymphatic dissemination before hematogenous spread to the lungs, liver, or bones. Notably, patients with hypermetabolic retroperitoneal and mediastinal lymphadenopathy experienced more rapid disease progression and higher mortality rates.35 Collectively, these findings highlight the critical importance of accurate staging to guide treatment selection, enabling identification of patients who may benefit from curative-intent therapy while sparing those with advanced disease from unnecessary morbidity.

Most professional societies agree that assessment for metastatic spread in bladder cancer is optimally performed with CT urography for evaluation of the upper urinary tract, while FDG PET/CT is generally considered superior for the detection of distant metastatic disease. However, despite its potential advantages, the routine use of FDG PET/CT has not yet been formally incorporated into major clinical guidelines.36,37 As a result, its clinical utility in this setting has been extensively investigated.

The majority of studies assessing FDG PET/CT in metastatic bladder cancer have been retrospective. In one of the largest series to date, involving 711 patients with invasive bladder cancer, FDG PET/CT identified previously unsuspected distant metastases in 9 % of cases, leading to a change in management from curative to palliative intent.35 Several additional studies have reported high diagnostic performance, with sensitivities ranging from 87 to 95 % and specificities from 78 to 90 %, often exceeding those of conventional CT imaging.29,38, 39, 40, 41, 42, 43 Conversely, other investigators have suggested that FDG PET/CT provides only limited incremental benefit over CT for detecting lymph node metastases beyond the pelvis and may have a restricted role in initial staging.44, 45, 46

Platinum-based neoadjuvant or induction chemotherapy is a standard component of treatment for patients with muscle‑invasive bladder cancer and selected patients with locally advanced disease, with the goal of downstaging tumors prior to radical cystectomy. Randomized trials have demonstrated that cisplatin-based neoadjuvant chemotherapy confers an absolute 5‑year overall survival benefit of approximately 8 % compared with surgery alone, supporting its routine use in eligible patients.47,48 However, outcomes after chemotherapy are heterogeneous. Patients with persistent metastatic disease following induction chemotherapy have a poor prognosis and derive little to no survival benefit from consolidative radical cystectomy, whereas patients with chemotherapy‑resistant but bladder‑confined disease may benefit more from early surgical intervention rather than from continued systemic therapy.49,50 Accurate assessment of response to systemic therapy in bladder requires a comprehensive approach incorporating clinical, pathologic and radiologic parameters. Traditionally, treatment response has been evaluated using size-based criteria, such as RECIST 1.1 and iRECIST,51,52 and recently with multiparametric MRI and VI-RADS scoring system.53

FDG PET/CT has emerged as a promising functional imaging modality for response assessment in this setting. A meta-analysis of five studies reported a pooled sensitivity of 0.84 (95 % CI, 0.72–0.91) and specificity of 0.75 (95 % CI, 0.59–0.86) for FDG PET in predicting tumor response following neoadjuvant or induction chemotherapy.54 In subgroup analyses, FDG PET/CT demonstrated a higher pooled sensitivity of 0.94 (95 % CI, 0.85–0.98) for distinguishing clinical responders from non‑responders, with a pooled specificity of 0.73 (95 % CI, 0.42–0.91), suggesting particular value in identifying patients unlikely to benefit from continued chemotherapy.

There is also evidence that FDG PET/CT performed during chemotherapy may provide prognostic information. In a retrospective study of 86 patients with bladder cancer and retroperitoneal lymph node metastases below the renal vessels, a complete metabolic response on interim FDG PET/CT after only a few cycles of cisplatin‑based chemotherapy was associated with significantly improved progression‑free survival compared with patients demonstrating partial or no metabolic response.55 These findings suggest that early metabolic changes may precede anatomic response and could potentially guide treatment escalation or early surgical referral.

Despite these encouraging results, FDG PET/CT has important limitations. Physiologic urinary excretion of FDG can obscure the bladder wall or mimic residual tumor activity in the bladder and distal ureters, reducing sensitivity for primary tumor assessment despite the use of delayed imaging, diuretics, or bladder irrigation. Furthermore, its sensitivity for detecting pathologic complete response remains limited, particularly in lymph nodes where inflammatory or treatment-related changes may metabolically mimic viable disease and lead to false-positive findings, especially shortly after chemotherapy.56 Conversely, small-volume residual disease or micrometastases may fall below the detection limit of PET, resulting in false-negative studies. Technical variability across institutions, including differences in acquisition protocols, acquisition times, timing of imaging relative to therapy, and methods for response quantification (e.g., SUV-based versus qualitative assessment) are variable that limit reproducibility and standardization. In addition, metabolic response on FDG PET/CT does not always correlate with pathologic response at cystectomy, underscoring the need for cautious interpretation and integration with anatomic imaging, clinical findings, and histopathologic confirmation. As a result, FDG PET/CT cannot currently replace surgical pathology for definitive response assessment. Larger prospective studies with standardized imaging protocols and response criteria are needed to better define the role of FDG PET/CT in monitoring and guiding therapy in patients undergoing neoadjuvant or induction chemotherapy for bladder cancer.

Despite definitive local therapy with radical cystectomy or TURBT, patients with muscle-invasive bladder cancer remain at substantial risk of relapse. Long-term outcome studies indicate that up to 50 % of patients develop distant recurrence following radical cystectomy, particularly those with advanced pathologic stage or lymph-node involvement at the time of surgery (20). Further, disease recurrence is associated with poor prognosis, with reported 5-year recurrence-free survival rates ranging from 58 % to 81 %.57

FDG PET/CT demonstrates high diagnostic accuracy for recurrence,58 with Alongi and colleagues reporting that FDG PET/CT not only detected recurrence but also provided prognostic information for both progression free and overall survival.59 Similarly, Zattoni et al. reported that FDG PET/CT outperforms CT and MRI, particularly in the detection of non-urinary tract recurrences, achieving a sensitivity of 94 % and a specificity of 79 % for recurrent urothelial carcinoma.60 Consistent with these findings, a meta-analysis of seven cohorts encompassing 603 patients demonstrated pooled sensitivity and specificity of 94 % and 91 %, respectively, for identifying recurrent or residual bladder cancer, frequently using delayed PET imaging protocols.61 Bacchiani et al. further corroborated these results, emphasizing the value of FDG PET/CT in clarifying indeterminate CT or MRI findings, detecting occult nodal or distant metastases, and frequently influencing subsequent clinical management.62 Overall, available evidence supports the use of FDG PET/CT as a valuable problem-solving tool in the restaging of bladder cancer. Although its routine use for surveillance is not yet universally recommended, FDG PET/CT provides important complementary information to conventional imaging in selected patients, and its clinical utility may be further enhanced by the integration of emerging non-FDG tracers and advanced quantitative imaging approaches into future diagnostic pathways.

FDG PET/CT is increasingly recognized not only for staging and restaging in bladder cancer but also as a prognostic or predictive imaging biomarkers.59,63 Beyond baseline risk assessment, the extent and distribution of FDG-avid disease on pretreatment PET/CT carry important prognostic implications. In particular, the presence of FDG-positive extrapelvic metastatic disease has been consistently associated with adverse outcomes.64 Moreover, changes in PET parameters following neoadjuvant chemotherapy or immunotherapy, such as reductions in SUV-based metrics or volumetric indices have been shown to correlate with pathological complete response and long-term clinical outcomes more accurately than anatomical imaging alone, supporting a role for FDG PET/CT in early response assessment and adaptive treatment strategies.63, 65, 66, 67

While conventional semi-quantitative parameters such as SUVmax and SUVmean correlate with tumor aggressiveness and unfavorable pathological features, their ability to predict outcomes is limited by technical and biological variability.68 Consequently, volumetric PET parameters have gained increasing attention for risk stratification. Metabolic tumor volume (MTV) and total lesion glycolysis (TLG), which integrate both the intensity and extent of metabolically active disease, provide comprehensive measures of whole-body tumor burden. Elevated baseline MTV appears to be an independent predictor of poorer overall and progression-free survival beyond established clinicopathological factors.59 In addition, FDG PET–based volumetric analysis may support radiotherapy planning by improving delineation of high-risk subvolumes for biologically guided dose escalation.69 Although challenges remain regarding standardization of acquisition, segmentation, and clinically relevant thresholds, accumulating evidence suggests that FDG PET-derived volumetric metrics may offer independent prognostic value in bladder cancer.60

Several PET radiotracers have been investigated in bladder cancer in effort to overcome the major limitation of FDG, namely its intense urinary excretion, which can obscure evaluation of the bladder and adjacent pelvic lymph nodes. Tracers such as ¹¹C-choline, ¹⁸F-choline, ¹¹C-acetate, and ¹¹C-methionine exhibit lower urinary activity and have therefore been explored for primary tumor detection, nodal staging, and assessment of distant metastases.70,71 Evidence from small, early-phase studies suggest that ¹¹C-acetate and ¹¹C/¹⁸F-choline can reliably visualize primary bladder tumors and demonstrate reasonable specificity for nodal disease; however, carcinoma in situ is often missed and specificity is reduced by post-treatment inflammation (e.g., after Bacillus Calmette-Guérin (BCG)).72, 73, 74 Some studies have also reported improved visualization of pelvic lesions compared with FDG, particularly for nodal disease.72,75 In a metanalysis by Kim et al. combining 10 studies (encompassing 282 patients) using 11C-acetate and 11C-choline PET/CT, pooled sensitivity and specificity for lymph node staging were 66 % and 89 %, respectively, reflecting the modest sensitivity but relatively high specificity of these tracers.7011C-Methionine has also been explored in small studies and case reports suggesting potential advantages over FDG, including higher lesion contrast and improved detection of select recurrences and metastatic lesions75,76; however, for its value lacks robust clinical validation. Overall, the diagnostic performance of these tracers in bladder cancer has been variable, with inconsistent sensitivity and specificity for nodal and distant staging. None demonstrate clear or reproducible superiority over FDG PET/CT in overall staging accuracy, prognostic stratification, or impact on clinical management, and their use remains limited.70 In addition, 11C PET tracers, unlike 18F labelled agents, have limited clinical applicability due to their short half-life (20 minutes) and the requirement for an on-site cyclotron.

More recently, an area of active and promising investigation has been the use of fibroblast activation protein inhibitor (FAPi) PET imaging. Fibroblast activation protein (FAP) is a type II transmembrane serine protease expressed on activated fibroblasts in proliferating tissue and, more prominently, on cancer-associated fibroblasts (CAFs) within the tumor microenvironment. CAFs play a key role in tumor progression, invasion, and metastasis, making FAP an attractive target for diagnostic imaging. Immunohistochemical studies support this biological rationale, demonstrating strong FAP expression in approximately 50–100 % of bladder cancer specimens, with uptake of [68Ga]Ga-FAPi PET correlating with immunohistochemical positivity.77 Compared with FDG, FAPi consistently demonstrates significantly higher tumor-to-background ratios (5.3–5.6 vs. 1.9–1.95; p = 0.001) and increased detection rates of metastatic disease (∼30 %), particularly in lymph nodes, lungs and peritoneum, including lesions missed by FDG or CT78, 79, 80 (Fig. 4). The improved performance of FAPi PET may be attributable to the abundant stromal component within bladder cancers, characterized by overexpression of fibroblast activation protein (FAP) in CAFs, which increases with tumor stage.81 A systematic review by Hagens et al., encompassing 22 retrospective studies and 68 patients with urothelial malignancies, further supports the favorable tumor-to-background contrast and superior lesion detectability of FAPi PET compared with FDG, particularly for primary tumors and nodal, peritoneal, and distant metastases. However, the included studies were predominantly small, retrospective, and methodologically heterogeneous, with limited histopathologic validation. In a small prospective study by Koshkin et al. FAPi PET-CT detected small nodal metastases in patients with both localized and metastatic urothelial cancer, that did not meet RECIST criteria by conventional imaging, leading to changes in clinical management.80 However, histopathologic validation is difficult to obtain. Unterrainer et al. is one of the few studies to include histopathologic confirmation of FAPi PET/CT findings.82 Overall, FAPI PET offers several theoretical advantages, including high tumor-to-background contrast, and simplified imaging protocols without the need for patient fasting. FAPi PET data consistently demonstrate higher lesion detectability than FDG and suggest that this agent may become a valuable tool for pelvic nodal and peritoneal staging, pending confirmation of its diagnostic performance in larger, standardized prospective studies.

Nectin-4 is a transmembrane cell-adhesion molecule that is highly overexpressed in urothelial carcinoma, with immunohistochemical expression reported in approximately 60–80 % of cases and minimal expression in most normal adult tissues.83,84 This tumor-selective expression profile supports its utility as both a prognostic biomarker and an attractive molecular imaging and therapeutic target. The clinical relevance of Nectin-4 is highlighted by the demonstrated survival benefit of enfortumab vedotin, a Nectin-4–directed antibody–drug conjugate that has shown superior overall survival compared with standard chemotherapy in advanced urothelial carcinoma.85 A major challenge of targeted therapies such as enfortumab vedotin is correctly identifying patients most likely to benefit. The expression of Nectin-4 in the primary tumor and metastases correlates with the amount of the drug present. Of the Nectin-4–targeted PET radiotracers developed to date, the ⁶⁸Ga-labelled bicyclic peptide ⁶⁸Ga-N188 exhibits high binding affinity and PET uptake with PET uptake strongly correlating with immunohistochemical Nectin-4 expression, underscoring its potential utility as a companion diagnostic for personalized treatment strategies.86,87 Preliminary first-in-human studies have shown favorable pharmacokinetics, accurate detection of Nectin-4 overexpressing lesions, and strong correlations between tracer uptake and tissue expression, achieving high sensitivity and specificity of 88.1 % and 100 %, at defined SUV thresholds with low nonspecific uptake in normal tissues s.88 Although diagnostic performance for primary tumors may be inferior to that of FDG, Nectin-4 PET imaging offers improved specificity for nodal disease and shows promise as a companion diagnostic to guide Nectin-4 targeted therapies and to explore mechanisms underlying therapeutic resistance87 (Fig. 5). Another antibody–drug conjugate target, trophoblast cell surface antigen 2 (TROP-2) has demonstrated meaningful clinical benefit in advanced and metastatic urothelial carcinoma, exemplified by the efficacy of the TROP-2–directed agent sacituzumab govitecan.87,89 Huang et al. reported favorable pharmacokinetics and promising diagnostic performance of [⁶⁸Ga]Ga-NOTA-T4, a novel TROP-2–targeted nanobody, in a pilot clinical study including patients with solid tumors.90 Thus, by enabling whole-body, noninvasive quantification of target expression, Nectin-4 PET and TROP-2 PET may have the potential to refine patient selection, support treatment monitoring, and improve understanding of resistance pathways.

Along similar lines, immune-PET imaging using radiolabeled antibodies or antibody fragments targeting the programmed death-1 (PD-1)/programmed death-ligand 1 (PD-L1) has shown a promising noninvasive approach to characterize the immune microenvironment in bladder cancer and to support patient selection for immune checkpoint inhibitor therapy. Radiolabeled monoclonal antibodies and antibody fragments, including ⁸⁹Zr-atezolizumab, ⁸⁹Zr-durvalumab, and ⁶⁴Cu-atezolizumab, have demonstrated the feasibility of whole-body PD-L1 imaging in early clinical and translational studies.91, 92, 93 PD-L1 PET has revealed marked inter- and intrapatient heterogeneity in tracer uptake across primary tumors and metastatic lesions, reflecting variable immune checkpoint expression that may not be captured by single-site biopsy.92 Beyond baseline patient stratification, PD-L1 PET imaging may enable longitudinal assessment of treatment-induced changes in target expression and facilitate early identification of immune-resistant disease. Despite this potential, clinical implementation is currently limited by restricted tracer availability, cost, and technical complexity, and its value remains speculative in bladder carcinoma.

Several other molecular targets have been investigated for their potential role in targeted imaging of bladder cancer, including human epidermal growth factor receptor-2 (HER2), vascular endothelial growth factor (VEGF), and prostate-specific membrane antigen (PSMA). HER2 exhibits variable expression in urothelial carcinoma and has emerged as a promising imaging target in selected patients, supported by clinical studies using radiolabeled antibodies, such as 68Ga-HER2 affibody PET/CT that demonstrate specific tumor uptake94,95; however heterogeneity of HER2 expression remains a limitation.96 VEGF, a key mediator of tumor angiogenesis, has demonstrated moderate to high expression in nodal metastatic bladder cancer with acceptable contrast compared to negative lymph nodes97 supporting its feasibility for targeted imaging approaches. In contrast, PSMA generally exhibit weak and inconsistent expression in urothelial carcinoma, particularly in metastatic lesions,97 which may limit their imaging utility.98 A systematic review by Tariq et al. reported variable PSMA expression in bladder cancer, with higher expression observed in low-grade urothelial tumors that diminished with increasing tumor grade.99 Similarly, Lin et al. reported that FDG PET outperformed PSMA PET in a prospective pilot study of patients with urothelial carcinoma, highlighting the weak PSMA staging in high-grade tumors.100

In summary, non-FDG PET tracers offer promising avenues to overcome key limitations of FDG PET/CT in bladder cancer. Among these, FAP and Nectin-4 targeted tracers appear particularly attractive due to their biological relevance and favorable imaging characteristics. Nevertheless, current clinical experience remains preliminary, and larger prospective, multicenter studies are required to establish its definitive clinical value.

Radiomics and artificial intelligence (AI) applied to PET imaging remain largely confined to the research domain, with growing but still limited clinical evidence.98 In bladder cancer, promising results using CT- and MRI-based radiomics have been reported for predicting depth of primary tumor invasion, local and systemic staging, and treatment response assessment.101, 102, 103, 104 Emerging efforts in PET imaging have also been explored across several malignancies, leveraging high-throughput extraction of quantitative features that reflect metabolic activity, intratumoral heterogeneity, and spatial patterns of tracer uptake to support machine learning–based prediction of tumor staging and prognosis.105, 106, 107 However, technical challenges including urinary tracer excretion, variability in tumor segmentation, small and heterogeneous cohorts, and lack of standardized radiomic pipelines limit reproducibility and generalizability. Consequently, AI-driven PET radiomics should currently be regarded as a promising research tool, requiring multicenter prospective studies, harmonized methodologies, and robust external validation before routine clinical implementation in bladder cancer imaging.