SEARCH

SEARCH BY CITATION

Keywords:

  • nonsmall cell lung cancer;
  • TNM staging;
  • magnetic resonance imaging-positron emission tomography;
  • integrated positron emission tomography-computed tomography;
  • whole body imaging

Abstract

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SOURCES
  8. REFERENCES

BACKGROUND:

The objective of this study was to assess whether coregistered whole brain (WB) magnetic resonance imaging-positron emission tomography (MRI-PET) would increase the number of correctly upstaged patients compared with WB PET-computed tomography (PET-CT) plus dedicated brain MRI in patients with nonsmall cell lung cancer (NSCLC).

METHODS:

From January 2010 through November 2011, patients with NSCLC who had resectable disease based on conventional staging were assigned randomly either to coregistered MRI-PET or WB PET-CT plus brain MRI (ClinicalTrials.gov trial NCT01065415). The primary endpoint was correct upstaging (the identification of lesions with higher tumor, lymph node, or metastasis classification, verified with biopsy or other diagnostic test) to have the advantage of avoiding unnecessary thoracotomy, to determine appropriate treatment, and to accurately predict patient prognosis. The secondary endpoints were over staging and under staging compared with pathologic staging.

RESULTS:

Lung cancer was correctly upstaged in 37 of 143 patients (25.9%) in the MRI-PET group and in 26 of 120 patients (21.7%) in the PET-CT plus brain MRI group (4.2% difference; 95% confidence interval, −6.1% to 14.5%; P = .426). Lung cancer was over staged in 26 of 143 patients (18.2%) in the MRI-PET group and in 7 of 120 patients (5.8%) in the PET-CT plus brain MRI group (12.4% difference; 95% confidence interval, 4.8%-20%; P = .003), whereas lung cancer was under staged in 18 of 143 patients (12.6%) and in 28 of 120 patients (23.3%), respectively (−10.7% difference; 95% confidence interval, −20.1% to −1.4%; P = .022).

CONCLUSIONS:

Although both staging tools allowed greater than 20% correct upstaging compared with conventional staging methods, coregistered MRI-PET did not appear to help identify significantly more correctly upstaged patients than PET-CT plus brain MRI in patients with NSCLC. Cancer 2013. © 2013 American Cancer Society.


INTRODUCTION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SOURCES
  8. REFERENCES

In an effort to improve diagnostic accuracy and to compensate for the drawbacks of individual imaging modalities, alternative hardware-based and software-based approaches to postprocessing alignment of each imaging device involve the fusion of different imaging techniques in a single scanner.1, 2 This and other similar efforts have been directed at synergistic effect of individual performance of computed tomography (CT), positron emission tomography (PET), and magnetic resonance imaging (MRI) beyond an ordinary additive effect of those individual components.

One of the most important indications for whole body (WB) imaging is the detection of metastases. PET-CT is the most popular imaging device for metastasis workup.3, 4 With the development of fast MRI techniques (fast, T2-weighted [T2W] spin-echo and 3-dimensional, gradient-echo sequences) and a rolling table platform, Currently, WB MRI also is being used for metastatic workup.5, 6

Currently, integrated PET-CT appears to be the imaging modality of choice for staging lung cancer.1, 2, 7 In a study comparing diagnostic performance between PET-CT and WB MRI, both PET-CT and WB MRI appeared to provide acceptable accuracy and comparable performance for the staging of nonsmall cell lung cancer (NSCLC); however, for determining metastasis (M) category, each modality has its own advantages.8 Thus, we hypothesized that the incorporation of PET information into WB MRI findings would increase the proportion of patients with correct upstaging compared with findings based on PET-CT or WB MRI alone. Conversely, current lung cancer staging workup includes PET-CT and dedicated brain MRI with the knowledge that 18F-fluorodeoxyglucose (FDG) PET and PET-CT have limitations in detecting brain metastasis.9, 10 Specifically, in clinically high-stage lung cancers, brain CT or MRI should be added as a routine imaging method for detecting extrathoracic metastasis.7, 10 Therefore, the objective of the current study was to compare the clinical effectiveness of coregistered WB MRI-PET with that of PET-CT plus dedicated brain MRI in the staging of clinically resectable NSCLC.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SOURCES
  8. REFERENCES

This prospective study was approved by our institutional review board, and written informed consent was obtained from all patients. The study was registered as a randomized clinical trial with ClinicalTrials.gov (NCT01065415).

Patients

This study was carried out in a single tertiary referral hospital. All patients who had histologic or cytologic proof of NSCLC were approached, and those patients who had stage I, II, or IIIA disease (other than N2 lymph node status) on the basis of clinical staging (including physical examination, laboratory findings at admission, and enhanced thoracic CT scans) were included (Fig. 1).

thumbnail image

Figure 1. This flow chart illustrates how patients were enrolled and randomized to the 2 different study groups: coregistered whole body magnetic resonance imaging-positron emission tomography (WB MRI-PET) and positron emission tomography-computed tomography (PET-CT).

Download figure to PowerPoint

Patients were excluded if they had received previous treatment for cancer or other malignancy and if they had poor pulmonary function, a poor Eastern Cooperative Oncology Group performance status (grade 3 or 4), concurrent medical diseases that made them inappropriate for surgery, or contraindications to MRI-based or gadolinium-based contrast agents (serum creatinine level >2.0 mg/dL). Patients who were eligible for curative surgery were enrolled consecutively in this study by a chest radiologist who had 8 years of experience in lung cancer interpretation. Participants were randomly assigned to a preoperative imaging strategy of either coregistered MRI-PET (new modality) or WB PET-CT plus brain MRI (standard treatment) for the staging of NSCLC according to a computer-generated randomization list using a random permuted block design created by an independent statistician. This process was not blinded to patients or researchers.

Image Acquisition

The PET-CT imaging, dedicated brain MRI, and WB MRI examinations were performed within 5 days of each other (mean interval, 1.5 days; range, 0-5 days). In both the coregistered MRI-PET group and the PET-CT plus brain MRI group, we used the same PET platform.

PET-CT and dedicated brain MRI

The details of the imaging methods used in this study have been described previously.11 Patients received an intravenous injection of 5.5 megabecquerels (MBq)/kg of FDG and then rested for 60 minutes before scanning. The images were acquired with the use of a PET-CT device (mainly with a Discovery STe scanner; GE Healthcare, Milwaukee, Wis), which included a PET scanner (2.5 minutes per frame in a 3-dimensional mode) and a 16-slice CT scanner. CT was performed according to a standard protocol with the following parameters: 140 KeV; 30-170 mAs; pitch, 1.75; and section thickness, 3.75 mm (to match the PET section thickness). Immediately after performing the unenhanced CT, emission PET was achieved in the identical transverse field of view. Attenuation-corrected PET images (voxel size, 3.9 × 3.9 × 3.3 mm) using CT data were reconstructed by using a 3-dimensional, ordered-subsets expectation maximization algorithm (2 iterations, 20 subsets).

Dedicated brain MRI was conducted using a 3T MRI scanner (Achieva; Philips Medical Systems, Best, the Netherlands) with a standard head coil.9 Three sequences were used for imaging, including a T2W, axial, turbo spin-echo pulse sequence (repletion time, 3000 msec; echo time, 80 msec) with fat suppression; a fluid-attenuation, inversion-recovery (FLAIR) spin-echo sequence (repetition time, 11,000 msec; echo time, 125 msec; inversion time, 2800 msec); and a noncontrast-enhanced and a contrast-enhanced, T1W spin-echo sequence (repetition time, 500 msec; echo time, 10 msec). The contrast-enhanced sequence was obtained after a bolus injection of 0.2 mM/kg paramagnetic contrast agent (Dotarem; Guerbet, Charles de Gualle Cedex, France).

MRI with diffusion-weighted MRI for coregistered MRI-PET

All MR examinations were performed with a 1.5-T imager (Magnetom Avanto; Siemens, Erlangen, Germany) using surface array coils, which also are referred to as a total imaging matrix. In brief, thoracic MR images were obtained with T2W images of half-Fourier acquisition with single-shot, turbo spin-echo (HASTE) and spectral selection-attenuated inversion recovery (SPAIR) fat suppression (FS). The diffusion-weighted MR images were obtained using a single-shot, spin-echo, echo-planar imaging (EPI) sequence with SPAIR FS with β values of 0, 100, and 700 seconds/mm2. The entire brain was imaged with a T1W, 3-dimensional gradient sequence using the same slice thickness and the same transverse orientation that were to obtain PET images. These images were postprocessed to make fusion images (coregistered MRI-PET) of MRI and PET using the coregistration technique of AquarisNetstation (Terarecon, Inc., San Mateo, Calif). The brain, liver, and spine were scanned with additional organ-specific MRI sequences and orientations that were optimized for the detection of extrathoracic metastasis. It took approximately 40 minutes (including total scan time) to complete the MRI examination in each patient.

Image Analysis

PET-CT and dedicated brain MRI

One chest radiologist (with 7 years of thoracic CT interpretation experience and 3 years of WB MRI interpretation experience) and 1 nuclear medicine physician (with 7 years of PET/CT interpretation experience) jointly evaluated the integrated PET/CT and dedicated brain MRI images in consensus. Both interpreters were blinded to the clinical and pathologic results.

Tumor staging was performed taking into consideration the lesion size, involvement of the surrounding organs or chest wall, and the distance of the primary tumor from the carina. Lymph node stations were evaluated by allocating lymph nodes into 10 groups according to lymph node map definitions for lung cancer staging recommended by the International Association for the Study of Lung Cancer.12 Details of the methods used to interpret and detect lymph node metastasis have been described in previous studies.2, 8, 10 Abnormal focal FDG uptake accompanied by a corresponding anatomic alteration was considered to be indicative of metastasis.8 The presence of abnormal uptake was determined qualitatively by comparing the uptake of the lesions in question with that of a mediastinal blood pool in the thorax or that of liver uptake in the abdomen, bones, or the remaining soft tissues. At brain MRI, metastases were considered present when an enhancing nodule(s) or mass was observed in the brain parenchyma on T1W images.

Coregistered MRI-PET

Two chest radiologists (with 8 years and 5 years of experience in WB MRI interpretation, respectively) and 1 nuclear medicine physician (with 10 years of PET-CT interpretation experience), who were blinded to the clinical and pathologic results, jointly evaluated all coregistered MRI-PET images in consensus. Tumor staging was performed taking into consideration the lesion size, involvement of the surrounding organs or chest wall, and the distance of the primary tumor from the carina. Lymph node metastasis was considered present for all lymph nodes that were interpreted as metastatic on either MRI or PET/CT images.13 For extrathoracic metastases, the presence of a metastatic lesion also was determined using inclusive criteria; namely, discrete lesions with high signal intensity on T2W, fast-spin-echo images were considered metastatic, particularly if the lesions demonstrated substantial enhancement on T1W images (eg lesions in the brain, liver, and kidneys, in which normal tissues have FDG uptake). In addition, discrete lesions with abnormally high FDG uptake were regarded as metastatic if they were accompanied by corresponding morphologic alteration on MR images.8

Reference Standard

Patients with stage I, II, or IIIA NSCLC without mediastinal lymph node metastasis underwent thoracotomy with resection of the primary lung cancer. During thoracotomy for surgical tumor resection, surgeons dissected all visible or palpable lymph nodes in the surgical filed, irrespective of the size of the lymph nodes, to confirm pathologic lymph node status. For pathologic lymph node staging in patients who did not undergo surgery, mediastinoscopic biopsy, endoscopic bronchial ultrasonographic biopsy, or supraclavicular lymph node aspiration were performed. Metastatic staging was confirmed by using additional dedicated imaging or biopsy or follow-up.

Primary Outcome and Follow-Up Study

The primary outcome was the correct upstaging of NSCLC, which we defined as staging by either coregistered MRI-PET or PET-CT plus brain MRI that correctly helped identify higher TNM-classified lesions compared with clinical staging (resulting in an actual stage shift), leading to further biopsy or follow-up imaging studies rather than direct thoracotomy and consequent higher stage regrouping, resulting in stage-appropriate treatment and prognosis prediction. The results were defined as over upstaging when WB imaging helped identify lesions with higher TNM classification, thus leading to subsequent higher stage regrouping compared with clinical staging, that eventually proved to be benign on biopsy or other imaging modality studies. Likewise, the results were defined as under staging when WB imaging failed to detect lymph node metastases that later had malignant cells identified on biopsy at endoscopic bronchial ultrasonography or mediastinoscopy, or at surgery, or when extrathoracic lesions were detected on follow-up imaging studies within up to 6 months. Patients were followed at 3-month intervals for 6 months. When lung or extrathoracic lesions were detected on follow-up imaging studies, the lesions were regarded as metastatic (ie, grown from a microscopic metastatic focus). The possibility of an inflammatory nature of new lesions was excluded by noting the interval growth of target lesions on at least 2 consecutive imaging studies.

Statistical Analysis

Assuming that the increased rate of correct upstaging based on WB PET-CT plus brain MRI study, compared with conventional lung cancer staging, was 15%,14 coregistered WB MRI-PET was expected to increase the rate of correct upstaging up to 30% (at least a 15% increment) on the basis of our previous work, which was a single-arm study that compared the visual correlation of WB MRI-PET and integrated PET-CT results.15 By using a 2-sided significance level of 5%, we calculated that 121 patients per group would be required to achieve 80% power to detect this effect. On the basis of the estimated number of eligible patients managed at our institution during the planned recruitment period, in total, 300 patients were scheduled for inclusion in the study, allowing for dropouts.

For descriptive statistics, medians and interquartile ranges are presented for continuous variables because of non-normality, and frequencies with proportions are presented for categorical variables. For comparisons of patient characteristics between the 2 groups, the Mann-Whitney test was used for continuous variables, and the chi-square test or the Fisher exact test was used for categorical variables. A 95% approximated or exact confidence interval also was calculated for the difference in proportions of staging between the WB MRI-PET and PET-CT plus brain MRI groups.

RESULTS

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SOURCES
  8. REFERENCES

Of 409 patients who were assessed, 54 were excluded because they did not meet inclusion criteria from January 2010 through November 2011. Of the remaining 355 patients, 55 declined to participate. The remaining 300 patients were randomly assigned to the coregistered WB MRI-PET group (151 patients) and the PET-CT plus brain MRI group (149 patients). Excluding 8 patients in the WB MRI-PET group and 29 patients in the PET-CT plus brain MRI group who declined to participate after randomization or who were averse to the MRI procedure because of claustrophobia, we finally included 143 patients in the WB MRI-PET group and 120 patients in the WB PET-CT plus brain MRI group for analyses, because no data were available on the study endpoints for the patients who dropped out (Fig. 1). The patients who were excluded after randomization did not differ significantly in terms of age, sex, clinical tumor stage, or histologic subtype of NSCLC.

Patient characteristics are summarized in Table 1. Between the WB MRI-PET and PET-CT plus brain MRI groups, there was no significant difference in terms of age, sex, Eastern Cooperative Oncology Group performance status, smoking status, clinical stage at presentation, or histologic subtype of NSCLCs.

Table 1. Baseline Comparison of Patient Characteristics at Clinical Staging
 No. of Patients (%) 
CharacteristicsWB MRI-PET Group, n = 143WB PET-CT & Brain MRI Group, n = 120P
  • Abbreviations: CT, computed tomography; ECOG, Eastern Cooperative Oncology Group; IQR, interquartile range; MRI, magnetic resonance imaging; NOS, not otherwise specified; NSCLC, nonsmall cell lung cancer; PET, positron tomography imaging; WB, whole body.

  • a

    Mann-Whitney test.

  • b

    Chi-square test.

  • c

    Fisher exact test.

Age: Median [IQR], y63 [57-69]62 [56-68.5].596a
Sex  .136b
 Men85 (59.4)82 (68.3) 
 Women58 (40.6)38 (31.7) 
ECOG performance status  .250b
 0137 (95.8)111 (92.5) 
 16 (4.2)9 (7.5) 
Smoking status  .510b
 Never60 (42)49 (40.8) 
 Exsmoker63 (44.1)48 (40) 
 Current smoker20 (14)23 (19.2) 
Clinical stage at presentation  .494b
 IA46 (32.2)45 (37.5) 
 IB51 (35.7)35 (29.2) 
 IIA18 (12.6)20 (16.7) 
 IIB18 (12.6)10 (8.3) 
 IIIA10 (7)10 (8.3) 
Histologic features  .274c
 Adenocarcinoma91 (63.6)74 (61.7) 
 Squamous cell carcinoma43 (30.1)40 (33.3) 
 Large cell carcinoma2 (1.4)2 (1.7) 
 NSCLC NOS4 (2.8)1 (0.8) 
 Pleomorphic carcinoma1 (0.7)3 (2.5) 
 Adenosquamous carcinoma2 (1.4)0 (0) 

Staging Performance in the Whole Body MRI-PET and PET-CT Plus Brain MRI Groups

Lung cancer was correctly upstaged in 37 of 143 patients (25.9%) in the coregistered WB MRI-PET group and in 26 of 120 patients (21.7%) in the PET-CT plus brain MRI group (difference, 4.2%; 95% confidence interval, −6.1% to 14.5%; P = .426) (Fig. 2). The upstaged difference (4.2%) between the WB MRI-PET and PET-CT plus brain MRI groups was not statistically significant (Table 2). Lung cancer was over staged in 26 of 143 patients (18.2%) in the MRI-PET group and in 7 of 20 patients (5.8%) in the PET-CT plus brain MRI group (difference, 12.4%; 95% confidence interval, 4.8%-20%; P = .003) (Table 3). The cancer was under staged in 18 of 143 patients (12.6%) in the MRI-PET group and in 28 of 120 patients (23.3%) in the PET-CT plus brain MRI group (difference, −10.7%; 95% confidence interval, −20.1% to −1.4%; P = .022) (Table 4).

thumbnail image

Figure 2. These images illustrate the correct upstaging (final stage, T1aN2M1b) in a man aged 62 years with lung adenocarcinoma. (A) This coregistered magnetic resonance imaging-positron emission tomography (MRI-PET) image obtained at the level of the suprahepatic inferior vena cava reveals a primary tumor (18 mm in greatest dimension) with high fluorodeoxyglucose uptake in the right lower lobe. (B) This coregistered MRI-PET image obtained at the level of the splenic vein reveals a 23-mm left adrenal nodule (arrow) with no fluorodeoxyglucose uptake, confirming true-negative metastasis in the left adrenal gland. (C) This enhanced, T1-weighted brain image obtained with a whole body imaging technique clearly discloses a metastatic nodule (arrow) with high convexity of the left parietal lobe (correct upstaging was achieved by detecting an M1b lesion).

Download figure to PowerPoint

Table 2. Proportion of Patients With Correctly Upstaged Disease in the 2 Groups
 No. of Patients (%) 
Baseline Disease StageWB MRI-PET Group, n = 143PET-CT & Brain MRI Group, n = 120Percentage Point Difference
  • Abbreviations: CT, computed tomography; MRI, magnetic resonance imaging; PET, positron tomography imaging; WB, whole body.

  • a

    Chi-square test (95% confidence interval, −6.1 to 14.5; P = .426).

IA8 (5.6)5 (4.2)1.4
IB15 (10.5)7 (5.8)4.7
IIA6 (4.2)7 (5.8)−1.6
IIB5 (3.5)4 (3.3)0.2
IIIA3 (2.1)3 (2.5)−0.4
Total37 (25.9)26 (21.7)4.2a
Table 3. Proportion of Patients With Over Staged Disease in the 2 Groups
 No. of Patients (%) 
Baseline Disease StageWB MRI-PET Group, n = 143PET-CT & Brain MRI Group, n = 120Percentage Point Difference
  • Abbreviations: CT, computed tomography; MRI, magnetic resonance imaging; PET, positron tomography imaging; WB, whole body.

  • a

    Chi-square test (95% confidence interval, 4.8-20.0; P = .003).

IA6 (4.2)1 (0.8)3.4
IB8 (5.6)2 (1.7)3.9
IIA4 (2.8)1 (0.8)2.0
IIB5 (3.5)0 (0)3.5
IIIA3 (2.1)3 (2.5)−0.4
Total26 (18.2)7 (5.8)12.4a
Table 4. Proportion of Patients With Under Staged Disease in the 2 Groups
 No. of Patients (%) 
Baseline Disease StageWB MRI-PET Group, n = 143PET-CT & Brain MRI Group, n = 120Percentage Point Difference
  • Abbreviations: CT, computed tomography; MRI, magnetic resonance imaging; PET, positron tomography imaging; WB, whole body.

  • a

    Chi-square test (95% confidence interval, −20.1 to −1.4; P = .022).

IA8 (5.6)8 (6.7)−1.1
IB4 (2.8)11 (9.2)−6.4
IIA2 (1.4)6 (5)−3.6
IIB4 (2.8)2 (1.7)1.1
IIIA0 (0)1 (0.8)−0.8
Total18 (12.6)28 (23.3)−10.7a

Staging Performance in Terms of TNM Classification

Coregistered WB MRI-PET performed relatively better compared with PET-CT plus brain MRI in characterizing tumor extent (5 of 37 patients [13.5%] vs 0 of 26 patients [0%] correctly upstaged, respectively) and in detecting lung-to-lung metastases (3 of 37 patients [8.1%] vs 0 of 26 patients [0%] correctly upstaged, respectively). Both WB MRI-PET and PET-CT plus brain MRI enabled the depiction of additional lymph node metastases within the thorax (in the area of the N2 or N3 lymph node station: 12 of 36 patients [32.4%] and 8 of 26 patients [30.8%] correctly upstaged, respectively). Extrathoracic metastases were identified in the brain, bone, and many other organs. Correct upstaging by detecting extrathoracic metastases was more frequent in the PET-CT plus brain MRI group (15 of 26 patients; 57.7%) than in the MRI-PET group (13 of 37 patients; 35.1%), (Table 5).

Table 5. Distribution According to Tumor, Lymph Node, and Metastasis Classification of Correctly Upstaged Cancers of Various Causes and Diagnostic Confirmation Methods in Two Different Groups
 Coregistered WB MRI-PET Group, n = 37PET-CT & Brain MRI Group, n = 26
TNM ClassificationNo. of Patients (%)Method Used, No.No. of Patients (%)Method Used, No.
  1. Abbreviations: CT, computed tomography; EBUS, endoscopic bronchial ultrasonography; LN, lymph node; MRI, magnetic resonance imaging; PET, positron tomography imaging; US, ultrasound; VATS, video-assisted thoracoscopic surgery; WB, whole body.

Tumor status    
 Tumor size4 (10.8)Lobectomy, 4  
 Tumor invasion1 (2.7)Lobectomy, 1  
LN status    
 N13 (8.1)Lobectomy/pneumonectomy and LN dissection, 31 (3.8)Lobectomy and LN dissection, 1
 N26 (16.2)EBUS, 2; mediastinoscopic biopsy, 3; lobectomy and LN dissection, 17 (26.9)EBUS, 3; mediastinoscopic biopsy, 2; lobectomy and LN dissection, 2
 N36 (16.2)EBUS-biopsy, 1; mediastinoscopic biopsy; 1; US-guided aspiration, 41 (3.8)EBUS-biopsy, 1
Metastasis status    
 Lung-to-lung metastasis3 (8.1)Follow-up CT, 3  
 Pleural seeding1 (2.7)VATS biopsy, 12 (7.7)VATS biopsy, 2
 Bone only6 (16.2)Biopsy, 63 (11.5)Bone scan, 1; biopsy, 1; MRI, 1
 Brain only4 (10.8)Follow-up MRI, 47 (26.9)Follow-up MRI, 7
 Liver only  1 (3.8)Follow-up PET, 1
 Adrenal only  1 (3.8)Biopsy, 1
 Abdominal LN only1 (2.7)Follow-up PET-CT, 1  
 Muscle only1 (2.7)Biopsy, 1  
 Bone and liver1 (2.7)Follow-up MRI, 11 (3.8)Follow-up PET, 1
 Bone and brain  1 (3.8)Follow-up PET and MRI, 1
 Bone and abdominal LNs  1 (3.8)Follow-up PET and abdominal CT, 1

DISCUSSION

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SOURCES
  8. REFERENCES

We expected that coregistered WB MRI-PET would help increase the rate of correct upstaging for lung cancer compared with PET-CT plus dedicated brain MRI. However, our results demonstrated an incremental difference of 4.2% in correct upstaging with coregistered WB MRI-PET, and the difference was not statistically significant. Conversely, with coregistered MRI-PET, under staging decreased significantly compared with PET-CT plus brain MRI. Thus, coregistered MRI-PET may help to guide an appropriate treatment and to correctly predict prognosis by significantly reducing the number of under staged patients. Conversely, coregistered MRI-PET may cause unnecessary and invasive staging workups because of higher rates of over staging than PET-CT plus brain MRI.

MRI is advantageous for characterizing soft tissue malignancies, such as brain, bone, muscle, head and neck, breast, and liver primary tumors. In addition, advanced MRI techniques can provide molecular imaging information, including MR spectroscopy, diffusion-weighted MRI, and perfusion imaging data, without subjecting patients to additional radiation exposure.16, 17 MRI-PET is expected to be more accurate than PET-CT for tumor classification in primary tumors, whereas MRI is more advantageous than CT for characterizing soft tissue malignancies. For lymph node status, our preliminary study indicates that coregistered MRI-PET is superior to PET-CT alone, because it has enhanced sensitivity for preoperative lymph node staging in patients with NSCLC.13 For determining metastatic status, MRI-PET can be advantageous compared with PET-CT for identifying metastases, depending on the target organs.8, 16, 17

In the current study, coregistered MRI-PET correctly upstaged the tumor classification in 5 patients. This was possible because morphologic thoracic MRI information allowed better delineation of tumor size and better detection of mediastinal tumor extent than information from the CT component of PET-CT. A similar superiority of MRI-PET compared with PET-CT for tumor classification also was noted in the study by Schwenzer and colleagues.18 With regard to lymph node classification, MRI-PET appeared to be superior to PET-CT for the detection of lymph node metastasis and consequent correct upstaging by adding diffusion-weighted MRI information to PET information.13 For staging metastases, the expectation that MRI-PET would be better than PET-CT plus brain MRI for detecting extrathoracic metastasis was not met, probably because the anticipated advantage of MRI-PET was much mitigated with an offset effect by adding dedicated brain MRI in the PET-CT plus MRI group and because extrathoracic metastases other than to the brain were relatively small in number.

Sequential imaging and subsequent coregistration of MRI and PET would lead to much longer acquisition time, thus increasing patient movement and motion, misregistering artifacts, and deteriorating image quality. Since silicon avalanche photodetectors operable within an MR system were introduced,19 several prototypes of simultaneous WB MRI-PET have been installed in clinical settings.20 Recently, Schwenzer et al18 compared the performance of MRI-PET imaging in the staging of lung cancer with that of PET-CT as the reference standard and compared the quantification accuracy of a new WB MRI-PET system with corresponding PET-CT data sets. Those authors concluded that MRI-PET imaging of the lung ensured diagnostic image quality in the assessment of pulmonary masses. Similar lesion characterization and tumor stage were observed on both PET-CT images and MRI-PET images in most patients. In our randomized trial, we focused on evaluating the clinical benefit of coregistered WB MRI-PET for the correct upstaging of resectable NSCLCs to help guide an appropriate management plan.

Our study had several limitations. First, there was a difference between the 2 groups in the number of patients who dropped out. We do not believe this self-selection would have contributed to a bias in our study. Those patients who dropped out did not differ between the 2 groups in terms of variables (age, sex, clinical tumor stage, or histologic subtype of NSCLC) that affected the primary outcome. Moreover, we achieved our goal for the power of the study. The resulting sample sizes of 120 patients in the PET-CT plus brain MRI group and 143 patients in the MRI-PET group provided 82% power to detect a difference of 15% versus 30%, respectively, at a significance level <5%. Second, instead of the simultaneous acquisition and integration of imaging data from MRI and PET (as in so-called “simultaneous MRI-PET”), coregistered images from those 2 modalities were used for image interpretation. This may have lessened the performance of coregistered MRI-PET compared with simultaneously acquired and integrated MRI-PET. The coregistered images may have been particularly disadvantageous in the identification or interpretation of small lymph nodes, lesions adjacent to mobile organs, or lesions adjacent to the chest or abdominal wall.21, 22 Third, in our study, we used 1.5-T MRI for WB imaging. It is well known that scan time is further mitigated at 3,0 T with identical resolution.6 Finally, because of the dual arm nature of our randomized clinical trial, we could not verify that the patients who were under staged with PET-CT plus brain MRI were correctly staged with coregistered MRI-PET. However, according to a single-arm study,13 coregistered MRI-PET helped to significantly improve the sensitivity for detecting lymph node metastasis compared with PET-CT alone, whereas specificity was not significantly hampered.

In conclusion, although both staging tools provided greater than 20% upstaging capability than conventional staging, preoperative staging with coregistered MRI-PET does not appear to allow the identification of significantly more patients with NSCLC who have advanced-stage tumors, lymph nodes, or extrathoracic metastasis (correct upstaging) compared with PET-CT plus brain MRI. Coregistered MRI-PET may be helpful for guiding appropriate treatment and predicting prognosis by reducing under staging, but it may cause an unnecessary and invasive staging workup because of higher rates of over staging than PET-CT plus brain MRI.

FUNDING SOURCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SOURCES
  8. REFERENCES

This work was supported by the Clinical Research Development Program at Samsung Medical Center, Seoul, Korea (CRL 110011 and CRL 110312).

CONFLICT OF INTEREST DISCLOSURE

The authors made no disclosures.

REFERENCES

  1. Top of page
  2. Abstract
  3. INTRODUCTION
  4. MATERIALS AND METHODS
  5. RESULTS
  6. DISCUSSION
  7. FUNDING SOURCES
  8. REFERENCES
  • 1
    Lardinois D, Weder W, Hany TF, et al. Staging of non-small-cell lung cancer with integrated positron-emission tomography and computed tomography. N Engl J Med. 2003; 348: 2500-2507.
  • 2
    Shim SS, Lee KS, Kim BT, et al. Non-small cell lung cancer: prospective comparison of integrated FDG PET/CT and CT alone for preoperative staging. Radiology. 2005; 236: 1011-1019.
  • 3
    Blomqvist L, Torkzad MR. Whole-body imaging with MRI or PET/CT: the future for single-modality imaging in oncology? JAMA. 2003; 290: 3248-3249.
  • 4
    Antoch G, Vogt FM, Freudenberg LS, et al. Whole-body dual-modality PET/CT and whole-body MRI for tumor staging in oncology. JAMA. 2003; 290: 3199-3206.
  • 5
    Lauenstein TC, Goehde SC, Herborn CU, et al. Whole-body MR imaging: evaluation of patients for metastases. Radiology. 2004; 233: 139-148.
  • 6
    Schmidt GP, Baur-Melnyk A, Haug A, et al. Comprehensive imaging of tumor recurrence in breast cancer patients using whole-body MRI at 1.5 and 3 T compared to FDG-PET-CT. Eur J Radiol. 2008; 65: 47-58.
  • 7
    Alberts WM. Diagnosis and management of lung cancer executive summary: ACCP Evidence-Based Clinical Practice Guidelines (2nd edition). Chest. 2007; 132( 3 suppl): 1S-19S.
  • 8
    Yi CA, Shin KM, Lee KS, 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.
  • 9
    Marom EM, McAdams HP, Erasmus JJ, et al. Staging non-small cell lung cancer with whole-body PET. Radiology. 1999; 212: 803-809.
  • 10
    Lee HY, Lee KS, Kim BT, et al. Diagnostic efficacy of PET/CT plus brain MR imaging for detection of extrathoracic metastases in patients with lung adenocarcinoma. J Korean Med Sci. 2009; 24: 1132-1138.
  • 11
    Moon SH, Choi JY, Lee HJ, et al. Prognostic value of F-FDG PET/CT in patients with squamous cell carcinoma of the tonsil: comparisons of volume-based metabolic parameters [published online ahead of print February 6, 2012]. Head Neck. 2012.
  • 12
    Rusch VW, Asamura H, Watanabe H, Giroux DJ, Rami-Porta R, Goldstraw P. The IASLC Lung Cancer Staging Project: a proposal for a new international lymph node map in the forthcoming seventh edition of the TNM classification for lung cancer. J Thorac Oncol. 2009; 4: 568-577.
  • 13
    Kim YN, Yi CA, Lee KS, et al. A proposal for combined MRI and PET/CT interpretation criteria for preoperative nodal staging in non-small-cell lung cancer. Eur Radiol. 2012; 22: 1537-1546.
  • 14
    Maziak DE, Darling GE, Inculet RI, et al. Positron emission tomography in staging early lung cancer: a randomized trial. Ann Intern Med. 2009; 151: 221-228, W-48.
  • 15
    Yi CA, Lee KS, Kim B-T, Kwon OJ, Shim YM. Staging of non-small cell lung cancer with visual correlation of whole body MR imaging and positron emission tomography (MRI/PET): efficacy comparison with integrated PET/CT [abstract VC31-12]. Paper presented at: Radiological Society of North America 95th Scientific Assembly and Annual Meeting; November 29 to December 4, 2009; Chicago, IL.
  • 16
    Antoch G, Bockisch A. Combined PET/MRI: a new dimension in whole-body oncology imaging? Eur J Nucl Med Mol Imaging. 2009; 36( suppl 1): S113-S120.
  • 17
    Heusner TA, Kuemmel S, Koeninger A, et al. Diagnostic value of diffusion-weighted magnetic resonance imaging (DWI) compared to FDG PET/CT for whole-body breast cancer staging. Eur J Nucl Med Mol Imaging. 2010; 37: 1077-1086.
  • 18
    Schwenzer NF, Schraml C, Muller M, et al. Pulmonary lesion assessment: comparison of whole-body hybrid MR/PET and PET/CT imaging—pilot study. Radiology. 2012; 264: 551-558.
  • 19
    Pichler BJ, Kolb A, Nagele T, Schlemmer HP. PET/MRI: paving the way for the next generation of clinical multimodality imaging applications. J Nucl Med. 2010; 51: 333-336.
  • 20
    Ratib O, Beyer T. Whole-body hybrid PET/MRI: ready for clinical use? Eur J Nucl Med Mol Imaging. 2011; 38: 992-995.
  • 21
    Metser U, Golan O, Levine CD, Even-Sapir E. Tumor lesion detection: when is integrated positron emission tomography/computed tomography more accurate than side-by-side interpretation of positron emission tomography and computed tomography? J Comput Assist Tomogr. 2005; 29: 554-559.
  • 22
    Nogami M, Nakamoto Y, Sakamoto S, et al. Diagnostic performance of CT, PET, side-by-side, and fused image interpretations for restaging of non-Hodgkin lymphoma. Ann Nucl Med. 2007; 21: 189-196.