Abstract
Background:
Computed tomography with bone scans (CT-B) has been widely used for staging metastatic hormone-sensitive prostate cancer (mHSPC), but whole-body magnetic resonance imaging (WB-MRI) is increasingly adopted. This study compares WB-MRI and CT-B in detecting metastatic sites, disease classification (CHAARTED and LATITUDE), and treatment outcomes in mHSPC.
Methods:
This retrospective study included patients with mHSPC diagnosed between February 2017 and August 2023 at 2 UK NHS hospitals. Patients underwent baseline staging with either WB-MRI or CT-B. Data on demographics, disease extent, and treatment were analysed. Patients were stratified using CHAARTED and LATITUDE criteria. Survival outcomes were assessed using Kaplan-Meier and Cox regression analyses.
Results:
Among 203 patients (120 WB-MRI, 83 CT-B), WB-MRI identified higher rates of bone-only disease (47% vs 22%, P < .001), high-volume (49% vs 22%, P < .001), high-risk (47% vs 18%, P < .001), and de novo metastatic disease (91% vs 65%, P < .001), but lower lymph node–only metastases (10% vs 26%, P = .003) and prior radical treatment (surgery: 2% vs 13%, P < .001; radiotherapy: 7% vs 25%, P < .001). CHAARTED (HR 4.922, 95% CI: 1.937-12.507, P < .0001) and LATITUDE (HR 4.807, 95% CI: 1.674-13.809, P = .003) classifications independently predicted overall survival, with significant volume/risk differences only observed in WB-MRI (P < .001 and P = .001, respectively).
Conclusions:
Whole-body magnetic resonance imaging appears to enhance staging accuracy and risk stratification in mHSPC, potentially influencing treatment decisions. While limited by retrospective design, these findings suggest that WB-MRI may optimise management in mHSPC, warranting further prospective validation.
Keywords
Introduction
Prostate cancer is one the most common cancers affecting men worldwide with over a million cases each year. 1 The management of prostate cancer requires a multidisciplinary approach, with initial staging at diagnosis being of paramount importance. The management of metastatic low- and high-volume disease can vary.
One meta-analysis of 9 studies involving almost 10 000 patients with prostate cancer reported that disease volume was an important factor in deciding on the most appropriate treatment. It reported that there might be differences among the androgen receptor–targeting agents (ARTA) in the likelihood of maximising overall survival (OS) according to the disease volume, and that docetaxel chemotherapy has a better impact on OS in high-volume disease compared with low-volume disease. This was attributed to lower volumes of disease, potentially having fewer androgen receptor–independent cells, therefore being more responsive to ARTA as opposed to conventional chemotherapy. 2
The imaging modalities used in the initial staging pathway for patients with metastatic prostate cancer are crucial. Computed tomography (CT) imaging in combination with bone scans (CT-B) is the current gold standard imaging modality to stage these patients. 3 However, other imaging techniques, including whole-body magnetic resonance imaging (WB-MRI) or positron emission tomography (PET) imaging, have been increasingly used, with PSMA PET/CT being highly recommended as a single procedure in diagnosing and staging high-risk prostate cancer.4,5 While PSMA PET/CT is considered a valuable tool, it is important to acknowledge its limitations. For example, WB-MRI may offer advantages in certain scenarios, such as in men with ductal carcinoma, where PSMA PET/CT may have a higher false-negative rate. 6
The proPSMA trial studied the use of prostate-specific membrane antigen positron emission tomography (PSMA-PET) imaging in comparison to conventional CT-B imaging. It showed that PSMA-PET imaging offered 27% greater accuracy compared with CT-B imaging in addition to lower radiation exposure and reduced rates of equivocal findings. They also reported that PSMA-PET imaging went on to alter management for 27% of their patients compared with 5% in the conventional imaging arm. 7 Using MRI may also allow 25% of men to avoid biopsy and reduce rates of overdiagnosis or detection of insignificant disease.8,9
Local hospitals differ in the type of imaging modalities available to them, and this is not an uncommon issue faced by hospitals across the United Kingdom. There consequently may be discrepancies in the detection of disease and its subsequent management. Given the variation in diagnostic imaging used, we looked to assess the differences in rates of metastatic disease identification, CHAARTED volume/LATITUDE risk classification, as well as overall management and patients’ outcomes between 2 hospitals within the same locality, one with CT-B and the other with WB-MRI imaging available.
Methods
A cross-sectional retrospective comparative study was conducted among 2 cohorts of patients with mHSPC diagnosis confirmed within the same urology multidisciplinary meeting treated at 2 hospitals within the same National Health System (NHS) Trust in the United Kingdom, specifically the Queen Alexandra Hospital in Portsmouth and St Mary’s Hospital on the Isle of Wight, between February 2017 and August 2023. These were 2 different cohorts of patients, but they were discussed in the same urologic multidisciplinary team (MDT), by the same urologists and the same team of medical and radiation oncologists. The cohorts underwent different staging diagnostic investigations for the definition of mHSPC over the same period. The primary reason for this difference is that St Mary’s Hospital on the Isle of Wight, which belongs to the same NHS Trust as Queen Alexandra Hospital in Portsmouth, does not have nuclear medicine facilities. Consequently, WB-MRI is the preferred staging modality at St Mary’s to avoid requiring patients to travel to the mainland for bone scans. This practice allowed for a natural experiment to observe the impact of different staging modalities on patient outcomes.
Inclusion criteria for this study were patients with a confirmed diagnosis of mHSPC; managed at either Queen Alexandra Hospital in Portsmouth or St Mary’s Hospital on the Isle of Wight; who underwent initial staging with either WB-MRI or CT-B between February 2017 and August 2023; with confirmation of diagnosis within the same urology multidisciplinary meeting.
Patient lists were collated, and various information was retrieved, including Gleason scoring, dates of biopsy and diagnosis, extent of disease, CHAARTED and LATITUDE risk status and treatment details. The analysis excluded 7 patients who underwent choline or PSMA PET scans.
The CHAARTED high-volume criteria included the presence of visceral metastases or at least 4 bone lesions with at least 1 lesion beyond the vertebral bodies and pelvis. Low-volume disease, on the contrary, includes all other cases that do not fit these criteria as defined by the CHAARTED trial. 10 The LATITUDE high-risk criteria required at least 2 of the following features to be met: a Gleason score of 8 or more, at least 3 bone lesions and the presence of visceral metastases. 11
The study was registered as an audit at Portsmouth Hospitals University NHS Trust (Audit ID No. 5621 on 14 April 2023 – ‘Comparison of imaging modalities for initial staging of patients with metastatic hormone-sensitive prostate cancer’).
Statistical analysis
Descriptive statistics were used to analyse the clinical data, with percentages used for binary variables and medians used for continuous variables, along with their respective dispersion values. The χ2 test was performed to compare binary variables, and the Mann-Whitney test for continuous variables, with a significance level of P < .05 deemed acceptable.
Progression-free survival was defined as the time from the diagnosis of metastatic disease to the first objective evidence of radiographic disease progression or death, whichever occurred first, as assessed by clinicians according to the RECIST 1.1 criteria. Overall survival was calculated from the diagnosis of metastatic disease until death or the date of last follow-up, with patients who had not experienced any events at the time of analysis being censored. OS and PFS were estimated using the Kaplan-Meier method and reported as medians with a hazard ratio (HR) and 95% confidence interval (95% CI), and compared using a 2-sided log-rank test with a significance level of P < .05 considered acceptable. The median follow-up time was calculated according to the Kaplan-Meier reverse method.
The prognostic value of clinical variables was assessed through univariable analysis using Cox regression analysis for OS and PFS. A multivariable stepwise Cox regression analysis was performed on OS and PFS using clinical baseline prognostic factors, with 2 models based on CHARTEED and LATITUDE, and a P value <.05 used as the cutoff for factors identified in the univariable analysis. The results were reported as HRs with 95% CI.
Data extracted from electronic medical records and anonymised data were analysed using R Statistical Software (v4.2.3; R Core Team 2023).
Results
Patient characteristics
A total of 120 patients were diagnosed with mHSPC using WB-MRI, while 83 patients were diagnosed with CT-B imaging. Table 1 presents the main characteristics of the patients based on radiological staging with WB-MRI or CT-B imaging.
Characteristics of patients with mHSPC by WB-MRI or CT-B.
ADT: androgen deprivation therapy, ARTA: androgen receptor–targeted agent, CT-B: computed tomography and bone scan, mHSPC: metastatic hormone–sensitive prostate cancer, PSA: prostate-specific antigen, WB-MRI: whole-body magnetic resonance imaging.
Statistically significant P values in bold.
The following significant differences were observed between the 2 cohorts of patients. The median Gleason score at diagnosis was higher in patients staged with WB-MRI compared with CT-B scans (9 vs 8, P = .017). There was a higher prevalence of de novo metastatic disease in patients assessed with WB-MRI compared with CT-B scans (91% vs 65%, P < .001) as well as a higher prevalence of exclusively skeletal metastatic disease (47% vs 22%, P = .002). In contrast, patients staged with CT-B scan had a higher percentage of exclusively nodal metastatic disease (26% vs 10%, P = .003). Prostate-specific antigen value was not different among the 2 patients’ cohorts (P = .634).
More than twice the number of patients studied with WB-MRI were classified as high-volume according to the CHAARTED criteria, compared with those staged with CT-B imaging (49% vs 22%, P < .001). Similarly, according to the LATITUDE risk classification, patients staged with WB-MRI were classified as high-risk more than twice as often as those staged with a CT-B scan (47% vs 18%, P < .001).
Patients staged with CT-B imaging were more likely to have received radical prostatectomy (13% vs 2%, P < .001) and radical radiation therapy (25% vs 7%, P < .001) compared with patients staged with WB-MRI. A higher percentage of patients staged with WB-MRI received upfront androgen deprivation therapy (ADT) and docetaxel-based combination therapy (65% vs 45%, P = .004) and prostatic consolidation radiotherapy (26% vs 13%, P = .011). Conversely, a higher percentage of patients staged with CT-B scan received upfront ADT and ARTA (34% vs 18%, P = .01).
Patient characteristics according to CHARTEED and LATITUDE volume/risk classes
From the comparison of patients in the same CHARTEED volume or LATITUDE risk classes with WB-MRI or CT-B scan, the following significant differences emerged (see Table 2). Patients stratified into the CHAARTED high-volume category with WB-MRI showed a significantly higher percentage of exclusively skeletal metastatic disease (46% vs 11%, P = .011) and de novo metastatic disease (95% vs 78%, P = .045). Similarly, patients whose disease was classified as high-risk according to LATITUDE with WB-MRI showed a higher percentage of exclusively skeletal metastatic disease compared with those staged with CT-B scan (49% vs 7%, P = .0029), alongside a lower percentage of both lymph nodal and skeletal metastatic disease (42% vs 73%, P = .041).
Characteristics of patients with mHSPC staged by WB-MRI or CT-B according to CHAARTED volume and LATITUDE RISK.
ADT: androgen deprivation therapy, ARTA: Androgen receptor targeted agent, cRT: consolidation radiotherapy, CT-B: CT scan and bone scan, LHRH: lutenising hormone-releasing hormone, M: metastases, mHSPC: metastatic hormone-sensitive prostate cancer, PSA: prostate-specific antigen, rRT: radical radiotherapy, rSurgery: radical surgery, WB-MRI: whole-body magnetic resonance imaging.
Statistically significant P values in bold.
Patients with low-volume disease according to CHARTEED who were staged with WB-MRI showed a higher percentage of exclusively skeletal metastatic disease compared with those staged with a CT-B scan (49% vs 25%, P = .0053). In addition, in the LATITUDE low-risk disease group, a higher percentage of patients with de novo metastatic disease was observed in those studied with WB-MRI compared with a CT-B scan (87% vs 58%, P < .001).
Patients with high-risk disease according to LATITUDE staged with CT-B scan were more commonly seen to have undergone radical prostatectomy (13% vs 0%, P = .0052).
Patients with CHAARTED low-volume disease staged with CT-B scan more frequently received previous radical prostatectomy, and ADT and ARTA as upfront therapy compared with those staged with WB-MRI (15% vs 3%, P = .031% and 35% vs 16%, P = .0253, respectively). Conversely, patients with CHAARTED low-volume disease staged with WB-MRI more frequently received ADT and docetaxel combination therapy (67% vs 43%, P = .0069) and prostatic consolidation radiotherapy (44% vs 15%, P = .0001).
Survival and prognostic factors
With median follow-up not significantly different between the 2 cohorts of patients assessed with WB-MRI or CT-B imaging (23.4 months vs 22.4 months, P = .3), no significant differences were observed between the 2 populations in terms of PFS (61.5% vs 51.4%, P = .9) and OS (72.9% vs 78.8%, P = .325) at 5 years (see Figure 1).
Clinical outcome of patients with mHSPC staged by MRI and CT-B scan. (A) Progression-free survival. (B) Overall survival.
In the multivariate analysis of PFS (see Table 3), according to the CHARTEED model, the only independent prognostic factors that emerged were the de novo metastatic disease (HR: 0.465, 95% CI: 0.231-0.937, P = .0321) and the CHARTEED classification (HR: 3.206, 95% CI: 1.657-6.202, P < .0001). In the LATITUDE model, both de novo metastatic disease (HR: 0.427, 95% CI: 0.206-0.884, P = .002) and the LATITUDE classification (HR: 4.157, 95% CI: 2.07-8.330, P < .001) emerged as independent prognostic factors.
Univariate and multivariate analysis for PFS and OS by baseline clinical characteristics.
CI: confidence interval, cRT: consolidation radiotherapy, H: high, HR: hazard ratio, L: low, M: metastases, mHSPC: metastatic hormone-sensitive prostate cancer, OS: overall survival, PFS: progression-free survival, rRT: radical radiotherapy.
cRT was added to covariates as rRT was considered among baseline covariates.
Statistically significant P-values in bold.
In the multivariate analysis of OS (see Table 3), the CHARTEED classification (HR: 4.922, 95% CI: 1.937-12.507, P < .0001) and Gleason (HR: 1.837, 95% CI: 1.125-2.998, P = .014) resulted as independent prognostic factors in the CHARTEED model. In the LATITUDE model, the only independent prognostic factor that emerged was the LATITUDE classification (HR: 4.807, 95% CI: 1.674-13.809, P = .003), with Gleason retaining borderline statistical significance (HR: 1.794, 95% CI: 1.004-3.206, P = .048).
Clinical outcome of patients with disease volume/risk according to CHARTEED and LATITUDE
In the univariate analysis, according to CHAARTED disease volume or LATITUDE disease risk classification based on different radiological staging with WB-MRI or CT-bone scan (see Table 4 and Figure 2), although the prognostic value of both classifications was confirmed in the general population for 5-year PFS (P < .001 for both, respectively) and OS (P < .001 for both, respectively), significant differences between high and low disease volume according to CHAARTED criteria and high- and low-risk disease according to LATITUDE were observed only in patients studied with WB-MRI, both for PFS (P < .001 for both, respectively) and OS (P < .001 and P = .001, respectively).
Clinical outcome of patients with mHSPC staged by WB-MRI and CT-B imaging according to CHAARTED disease volume and LATITUDE disease risk.
CI: confidence interval, CT-B: computed tomography and bone scan, H: high, L: low, mHSPC: metastatic hormone–sensitive prostate cancer, OS: overall survival, WB-MRI: whole-body magnetic resonance imaging.
Statistically significant P-values in bold.
Clinical outcome of patients with mHSPC staged by MRI and CT-B scan according to CHAARTED disease volume and LATITUDE disease risk groups. (A) CHAARTED – Progression-Free Survival. (B) LATITUDE – Progression-Free Survival. (C) CHAARTED – Overall Survival. (D) LATITUDE – Overall Survival.
Discussion
This retrospective study aimed to highlight the strengths and weaknesses of current diagnostic tools for staging HSPC, not to alter clinical practice. Computed tomography and bone scans, though less sensitive, have been widely used in registrational studies for mHSPC, limiting external validity of drug approvals. In today’s practice, MRI and PET scans are more common, especially when treatment choices hinge on disease volume per CHAARTED criteria.
Our study provides a real-world comparison within the same centre, involving similar patients and the same MDT. Two different staging methods were used for mHSPC over the same period due to the Isle of Wight’s lack of nuclear medicine facilities, necessitating MRI instead of bone scans to avoid patient travel to the mainland. This unique situation allowed us to compare MRI with CT and bone scans effectively.
Considering these premises, the results of this study indicate notable differences between WB-MRI and CT-B in classifying high-volume or high-risk and de novo disease, according to CHARTEED and LATITUDE risk stratification, respectively, in patients with mHSPC, primarily due to the higher prevalence of bone metastases using WB-MRI. The key findings of this study include higher rates of bone-only disease, high-volume disease, high-risk disease, and de novo metastatic disease identified using WB-MRI compared with CT-B. Conversely, CT-B imaging showed higher rates of lymph node–only disease, and previous radical surgery or radiotherapy. Furthermore, the CHAARTED and LATITUDE classifications were identified as independent prognostic factors for OS, with significant differences observed between disease volume or risk only with WB-MRI. These findings may have clinical implications for the staging and management of patients with mHSPC related to the different disease stratification by the WB-MRI in contrast to CT-B. They are consistent with that reported by Padhani et al. 12 They reported that despite CT’s ability to assess the structure of bone disease and determine whether they are lytic or sclerotic in nature, MRI is superior in the detection of bone metastases and assessment of therapy response. They also report that some bony lesions may be missed by CT-B. This was further evidenced by a systematic review of 23 studies studying the role of CT-B in the detection of metastases in low-volume prostate cancer. 13 Bone scans using technetium Tc-99m tracers may not demonstrate the complete extent of bone disease present due to these tracers picking up osteoblastic activity associated with lytic lesions. Given that the CHAARTED criteria include at least 4 bone lesions with at least 1 lesion beyond the vertebrae and pelvis, in its determination of disease to be high-volume, imaging patients with WB-MRI can lead to more of them being deemed to have high-volume disease, as demonstrated in our analysis here. 14 This higher likelihood of detecting bone disease with WB-MRI may also explain why these patients staged by WB-MRI are less likely to receive curative interventions such as radical prostatectomy or radical radiotherapy.
As aforementioned, patients staged with CT-B imaging were more likely to receive triplet therapy with ADT, ARTA and docetaxel chemotherapy, compared with those who underwent WB-MRI, thus suggesting that management of the patient’s cancer was more likely dependent on the clinician’s choice rather than on differences in disease volume and risk stratification. However, it is important to note that current treatment choices made by clinicians are based on these prognostic factors, including the presence of de novo metastatic disease, disease volume, and the presence of visceral metastases. Cattrini et al 15 further explain how disease volume, the localisation of metastatic disease and the timing of metastatic disease are important potential factors guiding clinician choice of therapy for patients with mHSPC. This highlights the importance of accurate assessment of these factors before subsequent therapeutic choices.
Although the population staged with WB-MRI more frequently showed clinically unfavourable prognostic features such as higher Gleason score, de novo metastatic disease and a greater distribution of high-volume or high-risk disease according to CHAARTED or LATITUDE classifications, 5-year PFS and OS were not different to that of patients assessed with CT-B scan. However, in the multivariate analysis, the presence of de novo metastatic disease and LATITUDE stratification were shown to be prognostic markers for PFS. In addition, the Gleason score and CHAARTED and LATITUDE risk classifications emerged as independent prognostic factors of OS, which have been highlighted previously. The Gleason score has been reported to be a prognostic biomarker in early-stage disease, as conveyed by Aziz et al. 16 Although in castrate-resistant disease, Kawahara et al 14 have also reported poorer castrate-resistant prostate cancer–free survival and OS outcomes in their CHAARTED high-volume and LATITUDE high-risk patient cohorts. Our findings were echoed by those of Gravis et al 17 who reported poorer survival outcomes in their high-volume mHSPC patients with de novo metastatic disease. It is also important to note that only WB-MRI assessment of CHAARTED disease volume and LATITUDE risk stratification in this series confirmed a statistically significant prognostic value for both PFS and OS.
The limitations of this study include its retrospective nature and lack of detailed information on other factors such as performance status and comorbidities, which may have confounded the reported survival outcomes. In addition, we acknowledge that the lack of digital rectal evaluation and comprehensive risk class stratification represents a further limitation. Finally, the fact that these are 2 cohorts of patients with slightly different characteristics introduces a degree of selection bias; while we have attempted to account for these differences through multivariable analysis, there may be unmeasured confounders that influence our results.
Nonetheless, this homogeneous study cohort included more than 200 age-matched patients, and its real-life nature increases the applicability of these results. This study supports the use of appropriate and sensitive radiological staging of patients with mHSPC to avoid unnecessary initial overtreatment and allow accurate prognostic assessment and subsequent choice of the most appropriate upfront systemic treatment. These decisions can have an impact on cost with regard to the scans undertaken, but also if unnecessary therapeutic interventions are spared and over-treatment avoided. Moreover, from the patient’s perspective, opting for a single MRI appointment as opposed to 2 appointments for their CT and bone scans would be more convenient for them.
Conclusions
We observed significant differences between WB-MRI and CT-B in detecting high-volume or high-risk and de novo disease in patients with mHSPC, primarily due to higher detection of bone-only metastases. This discrepancy may guide clinicians decisions when offering different systemic therapies according to disease burden. While PSMA PET/CT is an increasingly valuable tool for staging high-risk prostate cancer, WB-MRI offers several potential advantages, particularly in specific clinical settings. Whole-body magnetic resonance imaging is more widely available and less costly than PSMA PET/CT, making it a more accessible option for many patients. Moreover, WB-MRI may have a lower false negative rate in certain histological subtypes. Therefore, WB-MRI remains a valuable alternative, especially in settings where PSMA PET/CT is not readily available or when ductal histology is suspected.
Footnotes
Ethical considerations
The study, titled ‘Comparison of Imaging Modalities for Initial Staging of Patients with Metastatic Hormone-Sensitive Prostate Cancer’ (Audit study ID No. 5621 – 14.04.23), was approved as a Trust audit by the audit team of Portsmouth Hospitals University NHS Trust, Portsmouth, United Kingdom. Ethics approval was not required in accordance with local and national guidelines for audit and quality service improvement projects within the NHS Trust in the United Kingdom. Similarly, written informed consent was not necessary as the guidelines for audit/service improvement projects in NHS UK Trusts permit the use of anonymised patient data. This study complies with the guidelines for human studies and was conducted ethically in accordance with the World Medical Association Declaration of Helsinki.
Consent to participate
In accordance with local and national guidelines for audit and quality service improvement projects within the NHS Trust in the United Kingdom, the requirement for informed consent to participate was waived by the audit team of Portsmouth Hospitals University NHS Trust, as anonymised patient data was used.
Consent for publication
Not applicable.
Author contributions
All authors contributed to the publication according to the ICMJE guidelines for authorship. The authors’ contributions are specified as follows: Study conception and design: Akash Maniam, Giuseppe L. Banna. Acquisition of data: Mona Ali Hassan, Shobana Anpalakhan, Naoko Atsumi, Shyamika Acharige. Data analysis: Marina Campione. Data interpretation: Utku Lokman, Hajra Iqbal, Tomasz Olejnik, Maja Uherek, Daniel Wilby, Richard Robinson, Joanna Buckley, Joanna Gale. Manuscript drafting: Giuseppe L. Banna, Shobana Anpalakhan. Manuscript review and approval: All authors.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
Declaration of conflicting interests
The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: GLB reported personal fees from Astellas, AstraZeneca, and Amgen outside the submitted work. SAn received support for travel for conference attendance. All other authors declare no competing interests and report no conflict of interest.
Data availability statement
The data that support the findings of this study are available upon request to the corresponding author. The data that support the findings of this study are not publicly available due to containing information that could compromise the privacy of research participants but are available from the corresponding author (GLB) upon reasonable request.