Following Dr. Gorin’s talk on the role of molecular imaging in renal masses, Dr. Bjurlin discussed the role of molecular imaging in bladder cancer. He began by discussing considerations for a variety of radiotracers including 18F-FDG, 11C-choline, 11C-acetate, 99mTc-MDP, and 18F-NaF. Each of these has advantages and disadvantages. In particular, 18F-FDG PET/CT has particular value to detect osteolytic bone lesions but has limited ability to detect osteoblastic lesions and to assess the local tumor or lymph nodes. Both 11C-choline and 11C-acetate have minimal urinary excretion but require onsite cyclotron which limits their availability. In terms of bony disease, 99mTc-MDP is widely available, however, it has poor spatial resolution and is limited to detecting osteoblastic disease, and may have false-negative results. In contrast, 18F-NaF has improved accuracy with high sensitivity for osteolytic, osteoblastic, and mixed bony lesions but has high false-positive rates for benign bony disease, does not detect visceral disease, and is expensive.
Dr. Bjurlin then focussed on the assessment of the primary bladder tumor, emphasizing that assessment with FDG is limited due to pooling in the bladder which may obscure the lesion. Many studies have sought to use adaptive protocols including hydration, forced diuresis, and other approaches to empty the bladder to improve the characteristics of FDG imaging. However, there is limited data to suggest that the use of this imaging approach is superior to conventional approaches including cystoscopy and computed tomography imaging.
In the short-term following TURBT, DR. Bjurlin highlighted data attempting to quantify residual tumor using FDG-PET/CT. Patients with residual tumors had higher levels of uptake as quantified by SUVmean, SUVmax, and the thickness of avid disease. However, this approach has not been adopted to guide re-resection decisions.
Considering a large number of studies of a variety of imaging approaches, the test characteristics of PET/CT imaging for primary tumor staging in bladder cancer vary widely. This heterogeneity, along with overall relatively poor test characteristics, has limited enthusiasm for this approach.
While MRI is not molecular imaging, Dr. Bjurlin highlighted that it has been used for local bladder cancer staging. This has now been formalized in the VI-RADS multiparametric scoring system. A recent review demonstrated relatively favorable test characteristics with sensitivity in the range of 75-95% and specificity in the range of 75-97%.
For patients with MIBC, accurate evaluation of nodal and distant staging is essential for prognostication and appropriate treatment recommendations. Conventional imaging using computed tomography is relatively limited though is currently widely used. Dr. Bjurlin suggested that PET/CT may be useful in this context. However, this remains controversial. The current literature suggests that PET/CT has relatively limited sensitivity (~56%) without meaningfully improved sensitivity (56% vs 40 or 60%, respectively) or specificity (92% vs 92 or 91%, respectively) compared to CT or MRI. In terms of nodal staging, many studies have assessed the performance of PET/CT. Due to differences in study design, there is a wide range of test performance characteristics.
Among patients with high-risk MIBC, Dr. Bjurlin highlighted prospective work which compared CT staging alone to PET/CT. The PET/CT demonstrated metastatic disease which was not apparent on CT in 47% of patients which resulted in changes in treatment for 27% of patients. A more recent, larger retrospective study from Germany (n=711) suggested that the use of FDG-PET/CT was associated with upstaging in 25.5% and downstaging in only 0.4%. This translated to a change in treatment for a meaningful proportion of patients (8 to 20%).
Dr. Bjurlin then highlighted that MRI may outperform CT based on anatomic characteristics. Combining PET molecular imaging with MRI may therefore provide improved accuracy, in theory. However, two small pilot studies have failed to support this hypothesis.
He emphasized that the limited data available to date should not change management. However, there is an ongoing multi-center randomized controlled trial to address the value of 18F-FDG PET/CT compared to CT alone in the staging of patients with muscle-invasive bladder cancer (NCT02462239). This trial will help to better delineate the role of 18F-FDG PET/CT in muscle-invasive bladder cancer.
Dr. Bjurlin then transitioned to discussing 11C radiotracers. These have minimal urinary excretion, thus making them more favorable for evaluation of the urinary tract including the bladder. A systematic review of 10 studies examining these modalities for lymph node staging suggested that the sensitivity was approximately 66% and specificity of approximately 89%. However, there was significant between-study heterogeneity with lower specificity among prospective studies. He, therefore, concluded that, overall, 11C-based PET performs similarly to FDG-based PET.
Moving from staging to the assessment of response to neoadjuvant chemotherapy, Dr. Bjurlin highlighted that FDG-PET/CT may better identify responders compared to conventional CT (95% vs 79%) though identification of complete response was similar (68% vs 63%). The sensitivity of FDG-PET/CT was somewhat higher (71% vs 64%) though specificity was identical (60% vs 60%). In addition to determining response at the end of NAC, Dr. Bjurlin discussed a scheme in which repeated FDG-PET/CT is used after two cycles of NAC with the plan to proceed straight to cystectomy for those who failed to respond. This may be beneficial to reduce delays to cystectomy for patients with chemotherapy-resistant tumors.
The bone is the most common site of bladder cancer metastasis, thus bone imaging is important in the management of these patients. While 99mTc is the most commonly used radiotracer, Dr. Bjurlin suggested that NaF may have potentially better characteristics. However, this agent is not covered by CMS and thus is not used in routine clinical practice.
Dr. Bjurlin then moved to the increasingly important role of immunotherapy in bladder cancer. As with response to neoadjuvant chemotherapy, there is a need to predict patients who will respond to anti-PD-(L)1 therapy. While immunotherapy is efficacy for many, there are a large number of patients who do not derive clinical benefit but may be at risk of significant toxicity. Therefore, there is a need to identify responders as early as possible. One such approach is to use zirconium-89-labeled atezolizumab. A small study recently published in Nature Medicine suggested that baseline imaging results with 89Zr-atezolizumab were associated with tumor response.
Finally, Dr. Bjurlin discussed current guidelines from the NCCN and American College of Radiology, which I have summarized in the table below:
In conclusion, Dr. Bjurlin suggested that there is increasing data for the use of FDG-PET/CT in patients with muscle-invasive disease though there is no role in patients with non-muscle invasive disease.
Presenter: M. Bjurlin, DO, MSc, FACOS, Urologic Oncology, University of North Carolina Health.
Written by: Christopher J.D. Wallis, Urologic Oncology Fellow, Vanderbilt University Medical Center, @WallisCJD on Twitter, during the AUA2021 May Kick-off Weekend May 21-23.