In MIBC, clinical to pathologic stage discrepancy is relatively common. Postoperative pathologic upstaging occurs in 20–25% of cases staged by conventional imaging (CT/MRI) [8, 9]. On the other hand, the use of [18F]FDG PET/CT (adopted by current guidelines [10, 11]) to assess lymph node status upstages patients in about 20% of cases [11] in the locoregional context. Therefore, a more accurate staging examination for this pathology is still required.

CT remains widely available and is routinely used for initial whole-body staging of bladder cancer, predominantly for nodal and distant metastasis assessment. For local T-staging (assessing the depth of wall invasion), CT is limited by poor soft-tissue contrast and cannot reliably distinguish between lamina propria invasion and muscularis propria invasion in many cases. Therefore, CT is unreliable in differentiating between stages Ta to T3a tumors, although it is useful for detecting invasion into the perivesical fat (T3b) and adjacent organs. For nodal staging, CT relies on size and morphology criteria. Reported accuracy for the detection of locally advanced tumors or nodal metastasis is variable, with studies showing moderate accuracy (for example, sensitivity/accuracy for nodal disease often < 70%, with substantial false negatives for normal-sized metastatic nodes). Thus, CT provides reasonable specificity for bulky nodal disease but limited sensitivity for micro metastatic or small metastatic nodes [12,13,14].

Magnetic resonance imaging has superior soft tissue contrast resolution compared with computed tomography and can evaluate post-biopsy reactions, as enhancement of the tumor occurs earlier than that of the normal bladder wall due to neovascularization. Multiparametric imaging standardized by the VI-RADS scoring system has substantially improved noninvasive local (T) staging of bladder cancer. Multiple meta-analyses and prospective studies have demonstrated that mpMRI with VI-RADS exhibits high performance in predicting muscle invasion, with pooled sensitivities and specificities often reported in the range of 80–90% (depending on the VI-RADS threshold used and reader experience). VI-RADS also improves inter-reader agreement relative to non-standardized MRI readings and is increasingly recommended for preoperative assessment and for response assessment after neoadjuvant therapy. Both CT and MRI may be used for assessment of local invasion by T3b disease, or higher, but they are unable to accurately diagnose microscopic invasion of perivesical fat (T2 vs. T3a) [15,16,17].

[18F]FDG PET/CT has an established role in whole-body staging because of its ability to detect distant metastatic disease; however, its role in locoregional staging requires careful consideration. Systematic reviews and meta-analyses looking at [18F]FDG PET/CT for pelvic lymph-node staging in bladder cancer report moderate diagnostic accuracy with relatively low pooled sensitivity (50% in some pooled analyses) but high specificity (> 90%), meaning that [18F]FDG PET/CT is reasonably specific for positive nodal disease but misses a substantial fraction of nodal metastases, particularly small or micrometastatic nodes. [18F]FDG PET/MRI hybrids have shown promising results for primary tumor detection and improved soft-tissue delineation in small series and may enhance diagnostic performance for local tumor assessment compared with PET/CT in some analyses. However, urinary excretion of [18F]FDG and post-TUR-BT inflammation are practical limitations that can obscure or mimic disease within the bladder and adjacent tissues. Overall, [18F]FDG PET/CT is more useful for detecting occult distant metastases and for post-TUR-BT restaging [18, 19].

The strengths and weaknesses of each current bladder cancer staging method are summarized in Table 3 [12,13,14,15,16,17,18,19,20].

Recent histological studies have shown that PSMA is expressed in the vasculature of bladder cancer [21,22,23]. Li et al. evaluated PSMA expression in tissue from TUR-BT or RC of patients with MIBC and identified a correlation between the degree of PSMA expression and other clinicopathological characteristics, such as metastasis, clinical stage, invasive stage, and Ki-67 index [24]. Schreiber et al. also evaluated immunohistochemical PSMA-staining in 89 urothelial cell carcinoma samples and identified a correlation with disease progression, but not with recurrence [25]. In this way, PSMA emerges as a potential predictive or prognostic biomarker in bladder cancer, and cases that capture it more intensely may have characteristics that have not yet been identified, making them potential candidates for theranostic treatment.

Other studies were performed comparing PSMA and FDG PET/CT. Campbell et al. reported a prospective series of three patients and concluded that [18F]PSMA PET/CT provided adequate uptake for the detection of primary and metastatic lesions of urothelial carcinoma (UC), although FDG PET/CT had greater ligand uptake across lesions [26]. Lin et al. also compared PSMA and FDG PET/CT, with FDG PET/CT being more effective in detecting lymph node lesions [27]. On the other hand, in a retrospective analysis, Kudachi et al. reported moderate to high-grade 68 Ga-PSMA PET/CT expression when used for the surveillance of histologically proven UC, identifying multiple soft tissue, nodal, and skeletal metastases more effectively than FDG PET/CT [28]. Importantly, none of these studies used [18F]PSMA-1007, which improves the evaluation of the urinary tract due to the radiotracer’s predominant biliary excretion.

To the best of our knowledge, our study is the first to provide a comparative analysis of [18F]FDG and [18F]PSMA-1007 PET/CT in muscle-invasive bladder cancer (MIBC), highlighting the advantages and limitations of each radiotracer in detecting primary tumors, regional lymph node involvement, and distant metastases. The findings suggest that while both tracers demonstrate similar sensitivity for metastatic lesions, [18F]PSMA-1007 offers distinct advantages in evaluating primary bladder tumors due to its lower urinary excretion.

One of the biggest challenges of [18F]FDG PET/CT for MIBC imaging is the intense urinary elimination of [18F]FDG, which can obscure bladder lesions and limit accurate tumor delineation. Despite diuretic administration and delayed imaging, this issue persisted in two patients, leading to under-detection of bladder tumors with [18F]FDG. In contrast, [18F]PSMA-1007 shows predominant hepatobiliary clearance [29], demonstrating superior lesion visualization within the bladder, identifying lesions in five patients compared to only three with [18F]FDG. These findings reinforce the potential PSMA-targeted imaging and theranostic roles in MIBC, especially for primary tumor assessment.

Both radiotracers demonstrated comparable efficacy in detecting regional lymph node metastases, with three patients showing FDG-avid and PSMA-avid nodal involvement. The uptake values (SUVmax) were variable but overlapping between the two tracers, suggesting that both can be used effectively for nodal staging. However, one patient presented an FDG-avid pulmonary metastasis that was not detected by [18F]PSMA-1007. This probably occurred because of the proximity between the lesion and the spleen, an organ with high physiological [18F]PSMA-1007 uptake. In fact, other organs present limited PET/CT lesion detectability because of physiological tracer uptake, such as the liver for [18F]PSMA-1007 and the brain for [18F]FDG [30].

A key finding of our study is the substantial uptake of [18F]PSMA-1007 in certain MIBC lesions, suggesting potential theranostic implications. PSMA-targeted radionuclide therapy has been increasingly used in prostatic malignancies, and there are ongoing studies in multiple non-prostatic neoplasms, including patients with bladder cancer. The selectivity of PSMA-based ligands for tumor neo-vasculature suggests their potential as valid targets for vascular agents in cancer treatment [31]. Our results indicate that selected MIBC patients with high PSMA expression might benefit from such approaches. However, further investigations, including histopathological correlation and larger patient cohorts, are necessary to validate this hypothesis.

Despite the promising findings, our study has many limitations. The sample size was relatively small. Biopsies of all affected sites would have been very informative, but they were not possible for ethical reasons. The only patient who underwent RC presented with positive histopathological results in the bladder (classified as pT2a), indicating a false-negative result for both [18F]FDG and [18F]PSMA-1007. In contrast, all resected lymph nodes in this same patient were negative, in agreement with both PET exams. Most patients did not have an indication for radical cystectomy; therefore, confirmation of the metastatic nature of the labeled lesions was based on the evolution of the lesions in imaging exams performed throughout follow-up. Positive mediastinal lymph nodes, as detected by both [18F]PSMA-1007 and [18F]FDG, appeared to respond to neoadjuvant chemotherapy; however, false-positive results of both tracers due to inflammatory lymph nodes in resolution cannot be excluded. Future studies should include larger cohorts with histological validation to determine the precise role of [18F]PSMA-1007-targeted imaging and therapy in MIBC.

Studies involving [18F]FDG PET/CT and [18F]PSMA-1007 PET/CT are expensive. Therefore, this pioneering real-world study was conducted with a small cohort, aiming to represent a feasibility study. The analysis of this small cohort is fundamental for the design of larger studies. The proposed strategy has potential theranostic implications for immediate clinical implementation, particularly in translating the context of prostate cancer to bladder staging and treatment.

Larger cohorts with histological validation are needed to determine the precise role of [18F]PSMA-1007 imaging and therapy in MIBC. The results of the present study could potentially be enhanced by employing PET/MRI instead of PET/CT. PET/MRI combines the molecular imaging strengths of PET with the superior soft-tissue contrast and multiparametric functional capabilities of MRI, thereby improving the detection and characterization of bladder lesions, particularly in situations where urinary tracer excretion compromises accurate tumor visualization. This modality is especially advantageous for pelvic malignancies, where precise anatomical detail is critical, as highlighted in a recent international consensus guideline [32]. Accordingly, the use of [18F]PSMA-1007 with PET/MRI may further improve the diagnostic performance of the tracer and should be explored in future studies to validate and expand upon the findings of the present work.

The significant PSMA uptake observed in certain MIBC lesions in this study suggests promising theranostic applications, expanding the role of PSMA-targeted strategies beyond prostate cancer. In particular, the intense uptake of [18F]PSMA-1007 in primary bladder tumors and metastatic sites highlights the potential for using PSMA as a diagnostic and therapeutic target in selected patients with MIBC.

PSMA-targeted radionuclide therapy, such as 177Lu-PSMA, has already demonstrated clinical benefit in prostate cancer and is being explored in other PSMA-expressing malignancies, such as urothelial carcinoma. Because PSMA expression in bladder cancer appears to be linked to tumor neo-vasculature rather than the tumor cells themselves, this may offer a therapeutic avenue with less off-target toxicity.

Our real-world findings support the rationale for further investigation of PSMA-based theranostic approaches in MIBC, particularly for patients demonstrating high tracer uptake. Future studies integrating PSMA PET imaging with radionuclide-targeted therapy may help define a new precision medicine strategy for a subset of bladder cancer patients.