Welcome, this website is intended for international healthcare professionals with an interest in the treatment of Advanced Prostate Cancer. By clicking the link below you are declaring and confirming that you are a healthcare professional

You are here

Can Whole-body Magnetic Resonance Imaging with Diffusion-weighted Imaging Replace Tc 99m Bone Scanning and Computed Tomography for Single-step Detection of Metastases in Patients with High-risk Prostate Cancer?

European Urology, 1, 62, pages 68 - 75

Abstract

Background

Technetium Tc 99m bone scintigraphy (BS) and contrast-enhanced computed tomography (CT) or magnetic resonance imaging (MRI) of the pelvis and abdomen are universally recommended for detecting prostate cancer (PCa) metastases in cancer of all stages. However, this two-step approach has limited sensitivity and specificity.

Objective

Evaluate the diagnostic accuracy of whole-body MRI (WBMRI) as a one-step screening test for PCa metastases.

Design, setting, and participants

One hundred consecutive PCa patients at high risk for metastases prospectively underwent WBMRI, CT, and BS completed with targeted x-rays (BS/TXR) in case of equivocal BS. Four independent reviewers reviewed the images.

Measurements

This study compares the diagnostic performance of WBMRI, CT, BS, and BS/TXR in detecting PCa metastases using area under the curve (AUC) receiver operator characteristics. A best valuable comparator (BVC) approach was used to adjudicate final metastatic status in the absence of pathologic evaluation.

Results and limitations

Based on the BVC, 68 patients had metastases. The sensitivity of BS/TXR and WBMRI for detecting bone metastases was 86% and 98–100%, respectively (p < 0.04), and specificity was 98% and 98–100%, respectively. The first and second WBMRI readers respectively identified bone metastases in 7 and 8 of 55 patients with negative BS/TXR. The sensitivity of CT and WBMRI for detecting enlarged lymph nodes was similar, at 77–82% for both; specificity was 95–96% and 96–98%, respectively. The sensitivity of the combination of BS/TXR plus CT and WBMRI for detecting bone metastases and/or enlarged lymph nodes was 84% and 91–94%, respectively (p = 0.03–0.10); specificities were 94–97% and 91–96%, respectively. The 95% confidence interval of the difference between the AUC of the worst WBMRI reading and the AUC of any of the BS/TXR plus CT lay within the noninferiority margin of ±10% AUC.

Conclusions

WBMRI outperforms BS/TXR in detecting bone metastases and performs as well as CT for enlarged lymph node evaluation. WBMRI can replace the current multimodality metastatic work-up for the concurrent evaluation of bones and lymph nodes in high-risk PCa patients.

Take Home Message

Technetium Tc 99m bone scintigraphy and computed tomography of the pelvis and abdomen are universally recommended imaging techniques for detecting metastases in prostate cancer patients. Single-step whole-body magnetic resonance imaging with diffusion-weighted imaging is equivalent to these modes and may replace the standard approach.

Keywords: Prostate cancer, Magnetic resonance imaging, Metastases, Diagnostic accuracy.

1. Introduction

Less than 10% of prostate cancer (PCa) patients are diagnosed with metastases, and another 20–30% develop systemic disease after local treatment preferentially in the pelvic and lumbar lymph nodes and in the skeleton [1] and [2]. Nonlymphatic visceral metastases are detected in around 10% of castration-resistant PCa (CRPC) patients [3] . Accordingly, practice guidelines recommend technetium Tc 99m bone scintigraphy (BS) and contrast-enhanced computed tomography (CT) or magnetic resonance imaging (MRI) of the pelvis and abdomen to define the metastatic status in patients at high risk for metastases, that is, those with a high prostate-specific antigen (PSA) or a high Gleason score at diagnosis or a rapid PSA progression during follow-up [2], [3], [4], and [5].

The diagnostic accuracy of BS is poor. Its lack of specificity frequently leads to the indication of second-line examinations, most often plain x-rays [6] and [7]. Axial skeleton and whole-body MRI (WBMRI) have better sensitivity and specificity than BS [8] and [9]. In addition, MRI allows objective measurement of metastases and assessment of tumour response in bone [10] . The addition of diffusion-weighted imaging (DWI) to WBMRI enables the study of extraskeletal involvement, including lymph nodes and other soft-tissue metastases, without requiring intravenous contrast agents [11], [12], and [13]. Therefore, WBMRI/DWI positions itself as a potential single-step alternative to the combination of BS and CT or MRI in patients with high-risk PCa by improving the detection and measurability of metastases with the convenience of a single-step imaging technique.

This study is the first to compare prospectively the diagnostic accuracy of WBMRI/DWI against the standard combination of BS completed with targeted x-rays (BS/TXR) and CT for the detection of skeletal and visceral metastases in high-risk PCa patients.

2. Patients and methods

2.1. Patients

Patients were included between March 2007 and March 2010 if they presented at diagnosis with a Gleason score ≥8 and/or a PSA ≥20 ng/ml or with a PSA recurrence with a PSA doubling time (DT) ≤12 mo after radical treatment or when receiving androgen-deprivation therapy (ADT). Patients underwent Tc 99m BS/TXR, contrast-enhanced thoraco-abdomino-pelvic CT, and WBMRI, including DWI, within 60 d of inclusion. The local ethics committee approved the study, and informed consent was obtained from all patients (protocol CZOL446GB07T/Belgian registration number B40320096489).

2.2. Whole-body magnetic resonance imaging studies

All WBMRI studies were obtained from a single 1.5-T magnetic resonance magnet (Achieva 1.5T, Philips Healthcare, Best, The Netherlands) according to Lecouvet et al. [9] . Diffusion-weighted whole-body imaging with background body signal suppression (DWIBS) was performed according to Takahara et al. [14] . Technical parameters can be obtained from the corresponding author.

To assess bone metastases, patients were categorised separately by one senior musculoskeletal radiologist (R1) and one senior radiology fellow (R2) into one of the following categories: (1) normal or benign and (2) metastatic focal or diffuse according to previously reported criteria [8] and [10]. Readers were blinded to clinical and other findings. R1 and R2 also separately assessed the presence of enlarged lymph nodes on DWI multiplanar reconstruction (MPR) and T1-weighted images. Response Evaluation Criteria Solid Tumours (RECIST) 1.1 was used to define the threshold size criteria for lymph node enlargement (small diameter >10 mm), and anatomic areas were the same as those used at CT evaluation of nodes [15] . Lesions suggesting metastases in other organs were recorded as visceral metastases.

2.3. Tc 99m bone scintigraphy studies

BS was performed and categorised as previously described by senior and senior fellow nuclear medicine physicians into (1) normal/benign, (2) positive for bone metastases, or (3) equivocal (the image could not be confidently categorised into one of the former two subgroups, requiring additional imaging procedures) [8] . In this latter situation, targeted x-rays (TXRs) were directed to equivocal BS areas. The results of this sequential bone work-up (BS/TXR) were categorised into one of two categories: (1) negative or benign or (2) positive for bone metastases.

2.4. Computed tomography imaging

Standard assessment of lymph nodes and visceral metastases consisted of contrast-enhanced CT of the pelvis, abdomen, and thorax. CT images were reviewed separately by two senior radiologists (R3 and R4). Lymph nodes were considered positive if their short axis was >10 mm; anatomic locations were recorded following the international classification [15] . Lesions suggesting metastases in other organs (visceral metastases) were recorded.

2.5. Best valuable comparator

Patients were treated at the investigator's discretion and monitored for at least 6 mo. In the absence of a systematic histologic gold standard, a best valuable comparator (BVC) was used because bone and lymph node biopsies are not routinely performed to adjudicate positive imaging readings and systematic biopsies were refuted by the ethical committee. BVC consisted of the consensus review of all available initial imaging tests (BS/TXR and WBMRI for bone evaluation; CT and WBMRI for node evaluation); review of follow-up BS/TXR, CT, and MRI images; and clinical and biologic follow-up obtained after 6 mo [8] and [16]. This material was reviewed afterwards by all specialists (urologist, medical oncologist, nuclear medicine physicians, and radiologists involved in the study), who determined in consensus the final diagnosis for each patient in terms of bone metastases, enlarged lymph nodes, and “global” metastatic status (taking into account bone metastases and lymph nodes) at initial staging.

2.6. Data and statistical analyses

For comparison, “equivocal” results of BS were classified either as nonmetastatic (optimist BS reading) or metastatic (pessimist BS reading). Interobserver agreement was measured by Cohen's κ coefficient. Area under the curve (AUC) receiver operator characteristics were compared by Wald test from logistic regression models; specificity and sensitivity were compared using the McNemar test. All tests and confidence intervals (CIs) were two-sided, with α = 0.05. SAS v.9.2 statistical analysis software (SAS Institute, Cary, NC, USA) was used. Because of the exploratory nature of the study and the difficulty in estimating a theoretical benefit for WBMRI, the number of patients was not planned. Instead, interim analyses were performed every 25 patients. Because statistical significance had not evolved at an analysis of 75 and 100 patients, we decided to publish at that number.

3. Results

Patient characteristics are summarised in Table 1 . One hundred patients were prospectively enrolled: 44 at initial diagnosis because of a Gleason score ≥8 and/or a PSA ≥20 ng/ml and 56 with a rapid PSA recurrence (PSA DT ≤12 mo), including 35 receiving ADT. The mean age was 69 yr of age (range: 53–88).

Table 1 Basic characteristics of the patients

  No. (range) Metastatic by BVC
Total 100 68
Newly diagnosed: 44 24
 T stage T3b, T4 34
 PSA >20 ng/ml 39
 PSA, ng/ml, mean (95% CI) 32 (12–78)
 Gleason ≥8 24
PSA progression after local therapy 21 14
 PSA DT, mo, mean (95% CI) 6.7 (1.2–10.9)
 PSA, ng/ml, mean (95% CI)
PSA progression under ADT 35 30
 PSA DT, mo, mean (95% CI) 5.4 (1.2–11.6)
 PSA, ng/ml, mean (95% CI) 36.4 (17.5–122.5)

BVC = best valuable comparator; PSA = prostate-specific antigen; CI = confidence interval; DT = doubling time; ADT = androgen-deprivation therapy.

3.1. Comparison of whole-body magnetic resonance imaging and the combination of bone scintigraphy completed with targeted x-ray and computed tomography for assessing overall metastatic status

After BVC adjudication, 68 patients were considered metastatic ( Table 2 ). Readers agreed in 82 of 100 cases. Agreement for WBMRI readings was 94% (κ = 0.87; 95% CI, 0.77–0.97); for BS/TXR and CT, agreement for readings was 97% (κ = 0.93; 95% CI, 0.87–1.00). WBMRI readings agreed slightly better with BVC (κ = 0.80; 95% CI, 0.67–0.92) than BS/TXR and CT readings (κ = 0.74; 95% CI, 0.61–0.88), not reaching statistical significance (p = 0.08). The difference (95% CI; p) between the AUC for WBMRI and BS/TXR and CT (R3) was +5% (range: −1% to +12%; p = 0.12) for R1 and +1% (range: −7% to +8%; p = 0.88) for R2; the difference between the AUC for WBMRI and BS/TXR and CT (R4) was +1% (range: −1% to +14%; p = 0.08) for R1 and +2% (range: −6% to +10%; p = 0.61) for R2. The 95% CI of the difference between the AUC of the worst WBMRI reading and the AUC of any of the BS/TXR and CT lay between −7% and +10% (p > 0.1), thus within the noninferiority margin of ±10% AUC.

Table 2 Imaging results and receiver operator characteristics for predicting bone and lymph nodes metastases on best valuable comparator

    BVC for bone metastases and lymph bodes *          
    Negative

(n = 32)
Positive

(n = 68)
Total

(n = 100)
Sensitivity

%
Specificity

%
PPV

%
NPV

%
AUC

%
    No. No. No. (95% CI) (95% CI) (95% CI) (95% CI) (95% CI)
WBMRI (R1) Negative 31 4 35 94 96 98 88 95
  Positive 1 64 65 (86–98) (84–100) (92–100) (73–97) (91–100)
WBMRI (R2) Negative 29 6 35 91 91 95 83 91
  Positive 3 62 65 (82–97) (75–98) (87–99) (66–93) (85–97)
BS/TXR and CT (R3) Negative 31 11 42 84 97 98 74 90
  Positive 1 57 58 (73–92) (84–99) (91–100) (58–86) (85–96)
BS/TXR and CT (R4) Negative 30 11 41 84 94 97 73 89
  Positive 2 57 59 (73–92) (79–99) (88–100) (57–86) (83–95)

* Percentages as number of patients = 100.

BVC = best valuable comparator; PPV = positive predictive value; NPV = negative predictive value; AUC = area under the curve; CI = confidence interval; WBMRI = whole-body magnetic resonance imaging; BS/TXR = bone scintigraphy completed with targeted x-rays; CT = computed tomography.

Thirteen patients had visceral metastases (seven liver; five lung; one lung plus adrenal). Twelve of those patients had either bone metastases or enlarged lymph nodes detected at WBMRI and CT. Only one had isolated visceral metastases (lung plus adrenal) detected at WBMRI and CT.

In addition to metastatic lesions, CT and WBMRI images identified equally hepatic haemangiomas in nine patients, cystic renal lesions in nine patients, solid renal masses in three patients, adrenal tumours in three patients, aneurysms of the aorta or iliac arteries in four patients, and thyroid enlargement in two patients. Gadolinium injection was required to adjudicate two liver lesions and one adrenal mass. CT identified kidney stones > 5 mm in three patients that were not seen on WBMRI. WBMRI identified on DWIBS sequence one pulmonary lesion not seen on CT that was further confirmed to be small-cell cancer.

3.2. Comparison of the diagnostic performance of whole-body magnetic resonance imaging, bone scintigraphy, and bone scintigraphy completed with x-ray to detect bone metastases

After BVC adjudication, 51% patients had bone metastases ( Table 3 , Fig. 1 ). TXRs were requested in 14% patients because of equivocal BS findings. BS readers agreed in 98% of the cases (κ = 0.96; 95% CI, 0.90–1.00). WBMRI identified metastases in 5 of 44 patients (11%) with a negative BS and 5 of 14 patients (35.7%) with an equivocal BS ( Fig. 2 ). WBMRI readers R1 and R2 identified bone metastases in 7 of 55 (12.7%) and 8 of 55 (14.5%) patients with a negative BS/TXR, respectively. WBMRI significantly outperformed BS alone as well as BS/TXR. The difference (95% CI; p) between the AUC for the WBMRI and pessimistic BS scenario was +15% (range: +8% to +22%; p < 0.001) for R1 and +13% (range: +5% to +21%; p < 0.001) for R2; the difference between the AUC for the WBMRI and optimistic BS scenario was 11% (range: +5% to +17%; p < 0.001) for R1 and +9% (range: +3% to +15; p = 0.0088) for R2; and the difference between the AUC for the WBMRI and BS/TXR scenario was +8% (range: +3% to +13%; p = 0.003) for R1 and +6% (range: +1% to +12%; p = 0.05) for R2. The difference is the result of the increased sensitivity of WBMRI readings compared to BS/TXR (98% for R2 vs 86% for BS/TXR; p = 0.03) and for similar specificity (98% for both).

Table 3 Imaging results and receiver operator characteristics for predicting bone metastasis on best valuable comparator

    BVC for bone metastasis *            
    Negative

(n = 49)
Positive

(n = 51)
Total

(n = 100)
  Sensitivity

%
Specificity

%
PPV

%
NPV

%
AUC

%
    No. No. No.   (95% CI) (95% CI) (95% CI) (95% CI) (95% CI)
BS Negative 39 5 44            
Positive 1 41 42 Optimist& 80

(67–90)
98

(89–100)
98

(87–100)
83

(71–91)
89

(83–95)
Equivocal 9 5 14 Pessimist& 90

(78–97)
80

(66–90)
82

(70–91)
89

(75–96)
85

(78–92)
BS/TXR Negative 48 7 55   86 98 98 87 92
  Positive 1 44 45   (74–94) (89–100) (88–100) (76–95) (87–92)
WBMRI (R1) Negative 49 0 49   100 100 100 100 100
  Positive 0 51 51   (93–100) (93–100) (93–100) (93–100) (N/A)
WBMRI (R2) Negative 48 1 49   98 98 98 98 98
  Positive 1 50 51   (90–100) (89–100) (90–100) (89–100) (95–98)

* Percentages as number of patients = 100. & For optimist BS assessment, equivocal BS cases were considered nonmetastatic; for pessimist BS assessment, they were considered equivocal BS metastatic.

BVC = best valuable comparator; PPV = positive predictive value; NPV = negative predictive value; AUC = area under the curve; CI = confidence interval; BS = bone scintigraphy; BS/TXR = bone scintigraphy completed with targeted x-rays; WBMRI = whole-body magnetic resonance imaging.

gr1

Fig. 1 Whole-body magnetic resonance imaging (MRI) versus bone scintigraphy (BS) for bone metastasis detection in a 70-yr-old patient with newly diagnosed prostate cancer (prostate-specific antigen 142 ng/ml; Gleason score 9). (A) BS shows two foci of increased uptake (arrows). (B) Coronal T1 and (C) diffusion-weighted MRI images of the whole body confirm bone metastases within the right glenoid and left iliac bone (arrows). (D) The best valuable comparator confirms metastasis, with significant progression of bone involvement at 6 mo.

gr2

Fig. 2 Whole-body magnetic resonance imaging (MRI) versus false-negative bone scintigraphy (BS) for bone metastasis detection in a 65-yr-old patient with newly diagnosed prostate cancer (prostate-specific antigen 18 ng/ml; Gleason score 7 [4 + 3]). (A) BS (anterior-posterior and posterior-anterior views) shows no significant lesion. (B) Coronal T1 and (C) diffusion-weighted MRI images of the whole body confirm bone metastases within L3 and the left iliac bone (arrows).

3.3. Comparison of the diagnostic performance of whole-body magnetic resonance imaging and computed tomography to detect lymph node enlargement

After BVC adjudication, 44% of patients had lymph node enlargement ( Table 4 ). CT and WBMRI agreed in 77% of patients ( Fig. 3 ). Based on BVC, R3 and R4 missed enlarged lymph nodes by CT in 10 and 8 patients, respectively, and overdiagnosed lymph node enlargement in 3 and 2 patients, respectively. R1 and R2 failed to identify lymph node enlargement in 8 and 10 patients, respectively, by WBMRI and overdiagnosed lymph node enlargement in 1 and 2 patients, respectively. Inter-reader agreement was 91% for WBMRI (κ = 0.81; 95% CI, 0.68–0.93) and 93% for CT (κ = 0.85; 95% CI, 0.74–0.96). Agreement between CT and WBMRI by a pairwise comparison was substantial (κ range: 0.64–0.70). The difference (95% CI; p) between the AUC for WBMRI and CT (R3) was +4% (range: −4% to +13%; p = 0.35) for R1 and +1% (range: −7% to +9%; p = 0.83) for R2; the difference between the AUC for WBMRI and CT (R4) was +1% (range: −8% to +10%; p = 0.85) for R1 and −2% (range: −10% to +6%; p = 0.57) for R2. The AUCs of all readings range between 85% and 90%, the 95% CI of the worst MRI reader being within ±10% of that of any of the CT readers ( Table 4 ; both p > 0.1), suggesting that WBMRI is neither significantly superior nor significantly inferior to CT.

Table 4 Imaging results and receiver operator characteristics for predicting lymph node metastases on best valuable comparator

    BVC for lymph nodes *          
    Negative

(n = 56)
Positive

(n = 44)
Total

(n = 100)
Sensitivity

%
Specificity

%
PPV

%
NPV

%
AUC**

%
    No. No. No. (95% CI) (95% CI) (95% CI) (95% CI) (95% CI)
CT (R3) Negative 53 10 63 77 95 92 84 86
  Positive 3 34 37 (62–89) (85–99) (78–98) (73–92) (79–93)
CT (R4) Negative 54 8 62 82 96 95 87 89
  Positive 2 36 38 (67–92) (88–100) (82–99) (76–94) (83–95)
WBMRI (R1) Negative 55 8 63 82 98 97 87 90
  Positive 1 36 37 (67–92) (90–100) (86–100) (77–94) (84–96)
WBMRI (R2) Negative 54 10 64 77 96 94 84 87
  Positive 2 34 36 (62–89) (88–100) (81–99) (73–92) (80–94)

* Percentages as number of patients = 100.

BVC = best valuable comparator; PPV = positive predictive value; NPV = negative predictive value; AUC = area under the curve; CI = confidence interval; CT = computed tomography; WBMRI = whole-body magnetic resonance imaging.

gr3

Fig. 3 Whole-body magnetic resonance imaging (MRI) versus computed tomography (CT) for lymph node metastasis detection in a 68-yr-old patient with rising prostate-specific antigen (PSA) while undergoing androgen-deprivation therapy (PSA doubling time: 6.4 mo). (A) Coronal T1 and (B) diffusion-weighted MRI of the whole body show multiple abnormal lymph nodes within the iliac and lombo-aortic regions (arrows). (C) These enlarged lymph nodes are confirmed on the corresponding coronal reformatted abdomino-pelvic CT image (arrows). (D, E) Close-up views of the node measurements on MRI and CT images.

4. Discussion

Treatment guidelines universally recommend Tc 99m BS/TXR and CT or MRI to assess metastases at diagnosis in patients with a high PSA (>20 ng/ml) and/or high Gleason score (≥8) and in patients progressing after local treatment or ADT in cases of a short PSA DT [2], [3], [4], [5], and [17]. Because any universal one-step substitute must be effective across all PCa stages, this study enrolled patients fulfilling one of these descriptions. Based on BVC assessment, 68% of the patients were considered metastatic, including 51% in the skeleton and 44% in the lymph nodes—in line with our previous reports on the value of MRI in detecting bone metastases [8] and [9]. This study had only enrolled patients at high risk for metastases to compare two diagnostic strategies, not to evaluate the impact of WBRMI on diagnosis and treatment. The population may appear heterogeneous in terms of stage, but it is homogeneous in terms of diagnostic requirements. This approach was chosen, because the endpoints are purely diagnostic.

The multistep BS/TXR and CT approach has significant drawbacks. BS sensitivity and specificity are low, because BS detects bone deposits from osteoblasts, not cancer cells [18] . TXRs are often required to adjudicate unclear BSs, at additional expense and radiation [8] . MRI and WBMRI are superior to BS for detection of bone metastases, because they detect cancer cells in the bone before bone remodelling has occurred [9] and [19]. For the detection of enlarged lymph nodes and visceral metastases, CT and MRI perform similarly poorly compared to histologic diagnosis [3], [17], [20], [21], and [22]. It is expected, however, than in the future, MRI will progressively replace CT, because multiparametric MRI is already the gold standard for imaging the prostate, and awaited iron oxide particles should enhance the detection of enlarged lymph nodes [23] and [24].

WBMRI and positron emission tomography (PET) compete for single-step whole-body assessment of metastases and imaging of response to therapy in solid cancers. In lymphoma, breast, and lung cancers, WBMRI and DWI appear inferior or equivalent to fludeoxyglucose F 18 (FDG) [25] , whereas PET-FDG is less sensitive in PCa patients and in other cancers with sclerotic bone metastases [26] and [27]. Although whole-body imaging with 11C-choline or FDG is promising in PCa, WBMRI, including diffusion-weighted sequences, is superior to PET scanning for bone metastases [27] and [28]. Budiharto et al. have compared 11C-choline PET/CT and MRI/DWI with surgical lymph node staging in high-risk PCa patients and concluded that both have equivalently poor diagnostic performance [22] .

As expected from previous works, WBMRI outperformed BS/TXR in detecting bone metastases. WBMRI correctly identified bone metastases in a significant proportion of patients with negative BS/TXR imaging. Using RECIST 1.1 criteria, WBMRI and CT performed similarly for enlarged lymph node detection. These results suggest that WBMRI can replace the combination of BS/TXR and CT for a single-step metastatic work-up. A single-step WBMRI work-up presents several advantages. First, WBMRI offers the ability to quantify tumour load and response to therapy not only in soft-tissue metastases, which seldom occur in PCa but more importantly in bone, the most prevalent metastatic site [9], [10], and [29]. WBMRI offers the convenience of a single 30- to 45-min examination without contrast injection and irradiation—a significant advantage considering modern radioprotection concerns [30] . The cumulative irradiation of BS, TXR, and CT generates a dose effectively representing more that than several years of natural irradiation at each staging procedure.

The major limitation of this study is the definition of the gold standard because biopsies of bone metastases are not common practice and lymph node dissection (LND) is recommended only in patients in whom local treatment is discussed [3] . In particular, LND is not recommended at systemic recurrence or in CRPC patients, who are treated based on the imaging results [31] . Therefore, we have used a previously accepted BVC adjudication method [8] and [19]. In the present study, the gold standard (ie, the BVC) includes some clinical judgment or interpretation, and the result of the diagnostic test is used in the final adjudication process. Therefore, the gold standard and the diagnostic test under consideration are not independent of one another, generating an incorporation bias and an overestimation of the diagnostic accuracy of WBMRI. In the absence of histologic verification, there is no real alternative to the BVC, except maybe for a long-term analysis of progression. In the absence of such a prospective trial, it is impossible to evaluate the extent of such incorporation bias, and this will need to be tested. Assessment of bone metastases by WBMRI was (almost) perfectly identical to BVC, which limits the statistical comparisons for that endpoint but substantiates the previous arbitrary choice of bone MRI as a gold standard by several teams evaluating new nuclear medicine techniques targeting bone metastases [32] . In addition, the experience of the CT and MRI readers may have influenced the results. This potential bias will be addressed in an ongoing multicentre study.

This study does not provide an exhaustive incremental cost-effectiveness ratio (ICER) for WBMRI. In many countries, there is no specific billing code for WBMRI, and different centres may aggregate differently the cost of abdominal and skeletal MRI. Considering the wide variations in cost for CT and BS/TXR, every physician prescribing such imaging should ensure that the ICERs are acceptable and approved.

5. Conclusions

WBMRI is a promising, sensitive, and specific one-step technique for detecting bone metastases, enlarged lymph nodes, and visceral metastases in patients with high-risk PCa.


Author contributions: Frédéric E. Lecouvet had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Lecouvet, Tombal.

Acquisition of data: Lecouvet, El Mouedden, Coche, Danse, Jamar, Vande Berg, Omoumi, Tombal.

Analysis and interpretation of data: Lecouvet, El Mouedden, Collette, Coche, Danse, Jamar, Machiels, Vande Berg, Omoumi, Tombal.

Drafting of the manuscript: Lecouvet, Collette, Tombal.

Critical revision of the manuscript for important intellectual content: Lecouvet, El Mouedden, Collette, Coche, Danse, Jamar, Machiels, Vande Berg, Omoumi, Tombal.

Statistical analysis: Collette.

Obtaining funding: Lecouvet, Collette, Tombal.

Administrative, technical, or material support: None.

Supervision: Lecouvet, Tombal.

Other (specify): None.

Financial disclosures: Frédéric E. Lecouvet certifies that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None.

Funding/Support and role of the sponsor: This work was supported by a grant from the Fondation Saint Luc and by an educational grant from Novartis (CZOL446GB07T) to Dr Lecouvet and Dr Tombal. The contribution of Laurence Collette to this publication was supported by Fonds Cancer (FOCA) from Belgium. The sponsors of the trial had no role in the design and conduct of the trial, the analysis of the data, or the completion of the manuscript.

Acknowledgment

The authors thank N. deSouza, A. Baur, and F. Cornud for the input in the health-economic issue.

References

  • [1] G. Aus, S. Bergdahl, P. Lodding, H. Lilja, J. Hugosson. Prostate cancer screening decreases the absolute risk of being diagnosed with advanced prostate cancer—results from a prospective, population-based randomized controlled trial. Eur Urol. 2007;51:659-664 Crossref
  • [2] S.J. Freedland, E.B. Humphreys, L.A. Mangold, et al. Risk of prostate cancer-specific mortality following biochemical recurrence after radical prostatectomy. JAMA. 2005;294:433-439
  • [3] A. Heidenreich, G. Aus, M. Bolla, et al. EAU Guidelines on prostate cancer. Eur Urol. 2008;53:68-80 Crossref
  • [4] A. Briganti, N. Passoni, M. Ferrari, et al. When to perform bone scan in patients with newly diagnosed prostate cancer: external validation of the currently available guidelines and proposal of a novel risk stratification tool. Eur Urol. 2010;57:551-558 Crossref
  • [5] M.R. Smith, F. Kabbinavar, F. Saad, et al. Natural history of rising serum prostate-specific antigen in men with castrate nonmetastatic prostate cancer. J Clin Oncol. 2005;23:2918-2925 Crossref
  • [6] K. Nakanishi, M. Kobayashi, S. Takahashi, et al. Whole body MRI for detecting metastatic bone tumor: comparison with bone scintigrams. Magn Reson Med Sci. 2005;4:11-17
  • [7] H. Schirrmeister, A. Guhlmann, K. Elsner, et al. Sensitivity in detecting osseous lesions depends on anatomic localization: planar bone scintigraphy versus 18F PET. J Nucl Med. 1999;40:1623-1629
  • [8] F.E. Lecouvet, D. Geukens, A. Stainier, et al. Magnetic resonance imaging of the axial skeleton for detecting bone metastases in patients with high-risk prostate cancer: diagnostic and cost-effectiveness and comparison with current detection strategies. J Clin Oncol. 2007;25:3281-3287 Crossref
  • [9] F.E. Lecouvet, M. Simon, B. Tombal, J. Jamart, B.C. Vande Berg, P. Simoni. Whole-body MRI (WB-MRI) versus axial skeleton MRI (AS-MRI) to detect and measure bone metastases in prostate cancer (PCa). Eur Radiol. 2010;20:2973-2982 Crossref
  • [10] B. Tombal, A. Rezazadeh, P. Therasse, P.J. Van Cangh, B. Vande Berg, F.E. Lecouvet. Magnetic resonance imaging of the axial skeleton enables objective measurement of tumor response on prostate cancer bone metastases. Prostate. 2005;65:178-187 Crossref
  • [11] D.M. Koh, T. Takahara, Y. Imai, D.J. Collins. Practical aspects of assessing tumors using clinical diffusion-weighted imaging in the body. Magn Reson Med Sci. 2007;6:211-224 Crossref
  • [12] T. Komori, I. Narabayashi, K. Matsumura, et al. 2-[Fluorine-18]-fluoro-2-deoxy-D-glucose positron emission tomography/computed tomography versus whole-body diffusion-weighted MRI for detection of malignant lesions: initial experience. Ann Nucl Med. 2007;21:209-215 Crossref
  • [13] R.N. Low. Abdominal MRI advances in the detection of liver tumours and characterisation. Lancet Oncol. 2007;8:525-535 Crossref
  • [14] T. Takahara, Y. Imai, T. Yamashita, S. Yasuda, S. Nasu, M. Van Cauteren. Diffusion weighted whole body imaging with background body signal suppression (DWIBS): technical improvement using free breathing, STIR and high resolution 3D display. Radiat Med. 2004;22:275-282
  • [15] E.A. Eisenhauer, P. Therasse, J. Bogaerts, et al. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer. 2009;45:228-247 Crossref
  • [16] R. Venkitaraman, A. Sohaib, G. Cook. MRI or bone scan or both for staging of prostate cancer?. J Clin Oncol. 2007;25:5837-5838 author reply 5838–9 Crossref
  • [17] National Comprehensive Cancer Network. Clinical practice guidelines in oncology. Prostate cancer. NCCN Web site. http://www.nccn.org/professionals/physician_gls/f_guidelines.asp.
  • [18] H.E. Daldrup-Link, C. Franzius, T.M. Link, et al. Whole-body MR imaging for detection of bone metastases in children and young adults: comparison with skeletal scintigraphy and FDG PET. AJR Am J Roentgenol. 2001;177:229-236 Crossref
  • [19] R. Venkitaraman, G.J. Cook, D.P. Dearnaley, et al. Whole-body magnetic resonance imaging in the detection of skeletal metastases in patients with prostate cancer. J Med Imaging Radiat Oncol. 2009;53:241-247 Crossref
  • [20] G. Giannarini, G. Petralia, H.C. Thoeny. Potential and limitations of diffusion-weighted magnetic resonance imaging in kidney, prostate, and bladder cancer including pelvic lymph node staging: a critical analysis of the literature. Eur Urol.. 2012;61:326-340 Crossref
  • [21] A.M. Hovels, R.A. Heesakkers, E.M. Adang, et al. The diagnostic accuracy of CT and MRI in the staging of pelvic lymph nodes in patients with prostate cancer: a meta-analysis. Clin Radiol. 2008;63:387-395 Crossref
  • [22] T. Budiharto, S. Joniau, E. Lerut, et al. Prospective evaluation of 11C-choline positron emission tomography/computed tomography and diffusion-weighted magnetic resonance imaging for the nodal staging of prostate cancer with a high risk of lymph node metastases. Eur Urol. 2011;60:125-130 Crossref
  • [23] L. Dickinson, H.U. Ahmed, C. Allen, et al. Magnetic resonance imaging for the detection, localisation, and characterisation of prostate cancer: recommendations from a European consensus meeting. Eur Urol. 2011;59:477-494 Crossref
  • [24] R.A. Heesakkers, A.M. Hovels, G.J. Jager, et al. MRI with a lymph-node-specific contrast agent as an alternative to CT scan and lymph-node dissection in patients with prostate cancer: a prospective multicohort study. Lancet Oncol. 2008;9:850-856 Crossref
  • [25] C.A. Yi, K.M. Shin, K.S. Lee, et al. Non-small cell lung cancer staging: efficacy comparison of integrated PET/CT versus 3.0-T whole-body MR imaging. Radiology. 2008;248:632-642 Crossref
  • [26] H. Jadvar. Prostate cancer: PET with 18F-FDG, 18F- or 11C-acetate, and 18F- or 11C-choline. J Nucl Med. 2011;52:81-89 Crossref
  • [27] E. Even-Sapir, U. Metser, E. Mishani, G. Lievshitz, H. Lerman, I. Leibovitch. The detection of bone metastases in patients with high-risk prostate cancer: 99mTc-MDP planar bone scintigraphy, single- and multi-field-of-view SPECT, 18F-fluoride PET, and 18F-fluoride PET/CT. J Nucl Med. 2006;47:287-297
  • [28] W. Luboldt, R. Kufer, N. Blumstein, et al. Prostate carcinoma: diffusion-weighted imaging as potential alternative to conventional MR and 11C-choline PET/CT for detection of bone metastases. Radiology. 2008;249:1017-1025 Crossref
  • [29] I.F. Tannock, R. de Wit, W.R. Berry, et al. Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer. N Engl J Med. 2004;351:1502-1512 Crossref
  • [30] R. Heliou, L. Normandeau, G. Beaudoin. Towards dose reduction in CT: patient radiation dose assessment for CT examinations at University Health Center in Canada and comparison with national diagnostic reference levels. Radiat Prot Dosimetry. 2012;148:202-210 Crossref
  • [31] H.I. Scher, S. Halabi, I. Tannock, et al. Design and end points of clinical trials for patients with progressive prostate cancer and castrate levels of testosterone: recommendations of the Prostate Cancer Clinical Trials Working Group. J Clin Oncol. 2008;26:1148-1159 Crossref
  • [32] H. Schirrmeister, A. Guhlmann, J. Kotzerke, et al. Early detection and accurate description of extent of metastatic bone disease in breast cancer with fluoride ion and positron emission tomography. J Clin Oncol. 1999;17:2381-2389

Footnotes

a Department of Radiology, Cliniques universitaires Saint Luc, Université catholique de Louvain, Brussels, Belgium

b EORTC Headquarters, Statistics Department, Brussels, Belgium

c Division of Nuclear Medicine, Cliniques universitaires Saint Luc, Université catholique de Louvain, Brussels, Belgium

d Division of Medical Oncology, Cliniques universitaires Saint Luc, Université catholique de Louvain, Brussels, Belgium

e Division of Urology, Cliniques universitaires Saint Luc, Université catholique de Louvain, Brussels, Belgium

lowast Corresponding author. Department of Radiology, Cliniques universitaires Saint Luc, Avenue Hippocrate, 10/2942, B-1200 Brussels, Belgium.