Original article FDG-PET imaging in lung cancer: how sensitive is it for bronchioloalveolar carcinoma? Cecelia S Yap1, Christiaan Schiepers1, Michael C Fishbein2, Michael E Phelps1, Johannes Czernin1 1 Department
of Molecular and Medical Pharmacology, Ahmanson Biological Imaging Center/Nuclear Medicine, UCLA School of Medicine, AR-259 CHS, Los Angeles, CA 90095-6948, USA 2 Department of Pathology and Laboratory Medicine, UCLA School of Medicine, Los Angeles, California, USA Received 4 February and in revised form 07 April 2002 / Published online: 4 June 2002 © Springer-Verlag 2002
Abstract. While characterization of lung lesions and staging of lung cancer with fluorine-18 fluorodeoxyglucose positron emission tomography (FDG-PET) is an established clinical procedure, a lower diagnostic accuracy of FDG-PET for diagnosis and staging of so-called bronchioloalveolar carcinoma (BAC) has been reported. Therefore, the accuracy of PET for diagnosing and staging of BAC was investigated. We studied 41 patients eventually found to have adenocarcinoma with a bronchioloalveolar growth pattern who were referred for characterization or staging of lung lesions with wholebody FDG-PET between January 1998 and March 2001: there were 11 males (27%) and 30 females (73%), with a mean age of 66.0±10.9 (range =44–84 years). Patients were imaged using ECAT EXACT or HR+ systems. All patients had non-attenuation-corrected scans, while transmission data for attenuation correction were also available for 12 patients (29%). PET correctly identified BAC in 41 of the 46 (89%) lesions and 39 of the 41 patients (95%). By pathology, 25 patients (61%) were found to have unifocal or nodular lesions; this pattern was correctly identified by PET in 20 patients (80%) and by CT in 18 (72%). PET correctly identified 7 (44%) of 16 patients (39%) who had multicentric or diffuse BAC, and CT identified 11 (69%). Of the 35 patients whose lymph node status was verified pathologically, PET was correct in 27 (77%) and CT in 24 (69%). PET missed 67% of the rare tumors that had a pure BAC pattern with no invasive component. It is concluded that the diagnostic performance of whole-body FDG-PET is similar in most patients with lesions with a BAC pattern and in other non-small cell lung cancer types. PET is less accuJohannes Czernin (✉) Department of Molecular and Medical Pharmacology, Ahmanson Biological Imaging Center/Nuclear Medicine, UCLA School of Medicine, AR-259 CHS, Los Angeles, CA 90095-6948, USA e-mail:
[email protected] Fax: +1-310-2064899
rate in patients with rare BAC tumors that have no invasive component. Keywords: PET – Bronchioloalveolar carcinoma – Lung cancer – Adenocarcinoma – Pathology Eur J Nucl Med (2002) 29:1166–1173 DOI 10.1007/s00259-002-0853-y
Introduction Positron emission tomography (PET) with fluorine-18 fluorodeoxyglucose (FDG) identifies, stages and restages lung cancer [1, 2] and indeterminate lung nodules with a high accuracy [3]. Efforts to correlate the degree of FDG uptake in lung lesions with their biologic characteristics such as proliferative activity and tumor aggressiveness have produced mixed results [4, 5, 6] even though it is generally accepted that, among malignant tumors, less aggressive and slow-growing lesions exhibit a lower glucose metabolic rate than aggressive, rapidly growing lesions. Bronchioloalveolar carcinoma (BAC) is the least common type of bronchogenic carcinoma. It has a low incidence, occurring in only 2%–10% of all lung cancer patients. BAC has long been known as a well-differentiated adenocarcinoma [7, 8, 9] of the lung which has an extraordinary polymorphism in its clinical, radiological, and histopathologic presentations [10, 11]. BAC exhibits lepidic growth [7] and possesses unique clinicopathologic features that are different from those of non-BAC lung adenocarcinoma. Recent studies have suggested that the incidence of BAC is rising [12, 13]. The classification of BAC has undergone considerable changes. To address the problem of imprecise application of criteria for the diagnosis of BAC, the World Health Organization (WHO) in 1999 revised the definition of pure BAC to include only “non-invasive” lesions
European Journal of Nuclear Medicine Vol. 29, No. 9, September 2002
1167
[14]. In the previous WHO revision in 1981, BAC was defined as adenocarcinoma in which cylindrical tumor cells grow upon the walls of preexisting alveoli [15]. This is a much less rigorous definition than the current classification in which BAC is defined as an adenocarcinoma with a pure bronchioloalveolar growth pattern and no evidence of stromal, vascular, or pleural invasion. In the 1999 revision, BAC remains a subtype of adenocarcinoma but is now divided into three smaller categories: (1) non-mucinous (Clara cell/type II pneumocyte type), (2) mucinous (goblet cell type), and (3) mixed (mixed mucinous and non-mucinous; Clara cell/type II pneumocyte) or indeterminate. Previous studies suggested that FDG-PET could not reliably detect BAC [16, 17, 18, 19, 20]. These studies included few patients and utilized the older 1981 classification of BAC. Thus, these tumors included varying proportions of BAC components. However, no study to date has, in a larger group of patients, addressed the sensitivity of PET for detecting BAC according to the new WHO definition. The objectives of this study were fourfold. First, to determine the sensitivity of PET in identifying BAC using surgical pathology findings as gold standard. Second, to evaluate the ability of FDG-PET to discriminate unifocal or nodular from multicentric or diffuse disease. Third, to compare histologic with metabolic parameters, i.e., to evaluate the relationship between histologic tumor type and glucose metabolic activity. Lastly, to compare the accuracy of PET and CT for the staging of BAC.
Materials and methods Study population. Forty-one patients with the diagnosis of BAC underwent preoperative FDG-PET and CT imaging and surgical sampling or resection at our institution between January 1998 and March 2001. Patients who received any preoperative treatment between PET and surgery were excluded from the study. Because of its retrospective nature, this study was deemed exempt from the requirement of obtaining informed consent by the local ethics committee. The study group consisted of 30 women (73%) and 11 men (27%) whose age ranged from 44 to 84 years, with a mean of 66.0±10.9 years. Thirty-seven (90%) patients had one lesion identified by surgical pathology, three patients (7%) had two lesions, and one patient (3%) had three lesions. Thus, the study population of 41 patients had a total of 46 malignant lesions. Surgical removal or sampling of the 46 lesions was performed within an average of 28±24 days after the PET study (range 3– 106 days, median 22 days post PET). Overall, 50% of the lesions were removed by lobectomy, 24% by resection, 15% by wedge resection, and another 11% by pneumonectomy. FDG-PET image acquisition and reconstruction. Patients were instructed to fast for at least 6 h prior to PET imaging [21]. Serum glucose level at the time of FDG administration averaged 101±29 mg/dl. No intravenous insulin was administered. Emission scans of the body starting 45 min after intravenous administration
of approximately 555 MBq (15 mCi) of FDG were acquired for six to nine bed positions (6 min per bed position) for each patient. Patients were scanned using the Siemens-CTI EXACT or HR+ systems (CTI/Siemens, Inc., Knoxville, Tenn.). The reconstructed images had a resolution of 8–12 mm [22, 23]. Images for wholebody PET scans performed prior to July 2000 (n=31 patients) were routinely reconstructed using standard filtered back-projection without attenuation correction [24]. After July 2000, iterative image reconstruction algorithms were employed and both emission (4 min/bed position) and transmission (3 min/bed position) scans were performed in all patients so that whole-body attenuation-corrected images were available for evaluation [25, 26]. Hence, all 41 patients in the study population had non-attenuationcorrected images while 12 (29.3%) also had attenuation-corrected images and transmission data that allowed for SUV (standard uptake value) analysis. The acquired image sets were displayed in transaxial, coronal, and sagittal planes and were evaluated along with the 3-D volume images. PET image interpretation and data analysis. The interpretation of clinical PET images was conducted in both a blinded and a nonblinded manner. In the clinical setting, PET images were interpreted visually and correlated with available CT films or imaging reports and patients’ medical histories (non-blinded). In the blinded evaluation, PET images were interpreted by a nuclear medicine expert (J.C.) without any knowledge of the patients’ medical histories or other imaging reports. In the non-blinded evaluation, PET reports were generated based upon all the available clinical information (location, size, and number of lesions) derived from these reports. Each patient was assigned a lung cancer TNM stage based on the PET reports according to the revised International System for Staging Lung Cancer [27]. PET impression of the gross pathologic tumor feature was classified into three groups: no tumor, nodular (or unifocal), and multicentric (diffuse or multifocal). To ensure consistency in rating the FDG uptake within lesions, all PET images were reevaluated in a blinded fashion and scored visually by nuclear medicine experts (J.C., C.S.) using a scale for FDG uptake as follows: “no uptake” or normal physiologic uptake (same as background activity, or same as or less than mediastinal blood pool activity); mild uptake (greater than mediastinal blood pool activity); moderate uptake (greater than mediastinal blood pool activity but less than liver uptake); intense uptake (greater than liver uptake). Lymph node status was also assessed blindly. In the subgroup of patients in whom attenuation correction was performed (n=12), regions of interest (ROIs) were placed over the lesion or area of FDG accumulation. Semiquantitative analysis of the lesion was performed by calculating the maximum SUV. Corrections for lean body mass or partial volume effects were not applied. SUV is defined as the tissue concentration of FDG in the structure delineated by the ROI (kBq/ml) divided by the activity injected per gram body weight (kBq/g), as indicated below:
CT evaluation and data analysis. Only patients who had a CT scan within 3 months of PET were included. Chest CT scans were performed within an average of 21±26 days of PET. CT scans were performed either at our institution or outside facilities. Based on written clinical reports of the CT scan, each patient was assigned a lung cancer TNM stage according to the revised International System for Staging Lung Cancer [27]. CT impression of the
European Journal of Nuclear Medicine Vol. 29, No. 9, September 2002
1168 tumor gross feature was classified into three categories the same as those for PET (no tumor, nodular, and multicentric). Pathology review and data analysis. Pathology slides were retrieved and reviewed by an experienced pathologist (M.C.F.). The specimens were reclassified according to the recently revised WHO histologic and cytologic criteria for the diagnosis of BAC [14]. Gross pathologic features were noted wherever possible and tumor histology grades were categorized according to the conventional scheme as well, moderately, and poorly differentiated. Mucin production by the tumors was evaluated and confirmed via standard hematoxylin and eosin staining. Tumors were then categorized into three groups according to mucin content as mucin-producing, non-mucin-producing, or mixed. For the current evaluation, tumors had to be composed predominantly of mucinproducing cells to be considered mucinous. The extent of pure bronchioloalveolar component without invasion, within the tumor was evaluated microscopically through cut sections of the tumor and graded using the following scale: BAC1=1%–24% BAC component present, BAC2=25%–49%, BAC3=50%–74%, BAC4=75%–99%, and BAC5=100% (pure BAC). Of note, the recent, more stringent WHO classification results in a very small number of pure BACs. Statistical analysis. The relationship between tumor FDG uptake and pathologic features (histology grade and extent of BAC component) was calculated using linear regressions. The correlation between tumor size and FDG uptake was calculated using the Kruskal-Wallis test (a non-parametric test). Sensitivity was calculated in the usual fashion.
Results Pathology evaluation of lesions Pathologic findings of the 46 tumors are presented in Table 1. Eight (17%) lesions were BAC1 (1%–25% BAC component), 6 (13%) were BAC2 (26%–50% BAC comTable 1. Comparison of metabolic and histologic parameters: FDG uptake and histology
Histology
No.
ponent), another 8 (17%) BAC3 (51%–75% BAC component), 18 (40%) BAC4 (76%–99% BAC component), and 6 (13%) BAC5 (100% BAC component). Of the 46 lesions, 27 (59%) were located in the upper lobes and six (13%) were in the right middle lobe, while the remaining 13 (28%) were located in the lower lobes. The average lesion size as determined from the pathology specimen was 3.0±1.7 cm (median 2.7 cm, range 0.7–10 cm). FDG-PET findings and degree of FDG uptake in BAC Table 2 summarizes the results of the non-blinded clinical PET and pathologic findings for the 46 lesions found in the 41 patients. Of the 46 lesions, 41 (89%) were correctly identified by PET. The sensitivity of the blinded PET interpretation for lesion detection was 87%. A patient-based analysis shows that non-blinded PET correctly identified BAC in 39 of the 41 patients (95%). Both blinded and non-blinded visual analyses revealed no discrepancy between the attenuation-corrected and noncorrected images in the subgroup of 12 patients in whom whole-body transmission scans were available. Non-blinded visual analysis of PET images graded the majority of the tumors (n=29, 63%) as exhibiting intense FDG uptake, eight (17%) lesions as having moderate FDG uptake, four (9%) as showing mild uptake, and five (11%) as exhibiting no FDG uptake on PET. In the blinded evaluation of PET images, 29 (63%), six (13%), five (11%), and six (13%) of the tumors exhibited intense, moderate, mild, and no FDG uptake on PET, respectively. Thus, PET was false-negative in five of the 46 lesions (11%) in the non-blinded and in six of the 46 tumors (13%) in the blinded evaluation. However, both blinded and non-blinded evaluations identified BAC in 39 of 41 patients.
Ave. size (cm)
FDG uptake No FDG uptake
Mild
Moderate
Intense
Mucin production: Mucinous Non-mucinous
4 42
2.3±1.1 3.0±1.8
2 (50%) 3 (7%)
0 4 (10%)
2 (50%) 6 (14%)
0 29 (69%)
Histology grade: Well differentiated Moderately differentiated Poorly differentiated
5 25 16
2.5±1.2 2.5±1.1 3.7±2.3
1 (20%) 4 (16%) 0
0 3 (12%) 1 (6%)
1 (20%) 5 (20%) 2 (13%)
3 (60%) 13 (52%) 13 (81%)
BAC component: BAC1 BAC2 BAC3 BAC4 BAC5
8 6 8 18 6
4.1±3.0 3.1±1.2 2.4±1.1 2.9±1.1 1.8±1.0
0 0 1 (12.5%) 0 4 (67%)
1 (12.5%) 0 1 (12.5%) 2 (11%) 0
1 (12.5%) 2 (33%) 2 (25%) 3 (17%) 0
6 (75%) 4 (67%) 4 (50%) 13 (72%) 2 (33%)
European Journal of Nuclear Medicine Vol. 29, No. 9, September 2002
1169 Table 2. Patient characteristics, pathologic classification of BAC, FDG-PET imaging results, and CT classification of gross tumor features Pt. no.a
Age at PET (yr)
Sex
Time from PET to pathology (days)
BAC subclassb
Histology gradec
BAC grade (BAC1–5)
FDG uptake score (0–3)d
FDG SUVe
Gross tumor features by CTf
1a 2a 3 4 5 6 7 8 9 10 1b 11 12 13 14 15 16 2b 17 18 19 20 21 22 23 24a 25a 26 25b 27 24b 28 29 30 31 32 33 34 35 36 37 38 24c 39 40 41
67 70 75 71 50 45 75 77 81 70
F F M F F F M F F F F M F F F M
53 77 76 65 77 44 74 64 77 75
F M F F M F M M M F
45
F
67 55 73 45 77 59 60 59 66 62 70
F F F F F F M F F M F
84 69 52
F F F
Non-Muc Muc Non-Muc Muc Non-Muc Non-Muc Non-Muc* Non-Muc* Non-Muc Muc Non-Muc Non-Muc Non-Muc Non-Muc* Muc Non-Muc Non-Muc Non-Muc Non-Muc Non-Muc* Non-Muc Non-Muc Non-Muc Non-Muc Non-Muc Non-Muc Non-Muc Non-Muc Non-Muc Non-Muc Non-Muc Non-Muc Non-Muc Non-Muc Non-Muc Non-Muc Non-Muc Non-Muc Non-Muc Non-Muc* Non-Muc Non-Muc Non-Muc Non-Muc Non-Muc Non-Muc
Mod Well Mod Mod Mod Poor Mod Mod Mod Mod Mod Mod Mod Poor Well Mod Poor Mod Poor Poor Poor Poor Poor Poor Poor Poor Poor Mod Mod Mod Poor Well Well Mod Mod Mod Mod Mod Mod Mod Mod Poor Poor Poor Well Mod
3 5 5 5 5 1 3 4 4 1 2 2 3 3 4 4 4 1 1 1 1 1 1 2 2 2 2 3 3 3 3 4 4 4 4 4 4 4 4 4 4 4 4 4 5 5
0 0 0 0 0 1 1 1 1 2 2 2 2 2 2 2 2 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3 3
– – – Und Und – – – – 2.6 – – – 2.1 1.9 – 2.9 – – – – – – – – – – – – 4.5 – 6.4 – – – – 3.1 1.9 – – 3.1 – – – 1.8 –
D D D D D N N N N N
73 67 72 73 52 63
61 77 10 17 5 28 26 51 8 8 61 106 8 5 29 13 27 77 81 33 35 34 29 7 8 13 22 37 51 11 13 51 15 22 23 15 34 10 70 6 43 3 13 14 50 13
a 1a and 1b, 2a and 2b, 25a and 25b represent the two lesions found in each of the three patients (patients 1, 2, and 25); 24a, b, and c represent the three lesions in one patient (patient 24) b For BAC subclass: Muc, mucinous; Non-Muc, non-mucinous; Non-Muc*, predominantly non-mucinous with focal mucin positivity
D D N N N N N N N N D D N D D N N D D D N D D N N N D N N D N
c For histology grade: Well, well differentiated; Mod, moderately differentiated; Poor, poorly differentiated d For FDG uptake score: 0, no FDG uptake; 1, mild FDG uptake; 2, moderate FDG uptake; 3, intense FDG uptake e For FDG SUV: Und, undefined or not determined f For gross tumor features by CT: D, diffuse disease; N, nodular disease
European Journal of Nuclear Medicine Vol. 29, No. 9, September 2002
1170
Tumor size and degree of FDG uptake were significantly correlated (r2=0.77, P=0.04). SUV was determined for ten lesions in the 12 patients with transmission data (22% of the 46 lesions studied), yielding an average value of 3.0±1.4 and a range of 1.8–6.4. SUV was not defined in two patients because no lesions were found on visual analysis of the PET scans. The average SUV for mucinous (n=2) and non-mucinous (n=8) tumors were 2.25 and 3.33, respectively. Visual FDG uptake score of tumors stratified by mucin production, tissue grade or differentiation, and BAC component are shown in Table 2. PET and CT findings of tumor gross features and presentation Review of gross pathologic features identified nodular tumor lesions in 25 of the 41 patients (61%) and multicentric or diffuse disease in 16 (39%). Non-blinded PET, blinded PET, and CT correctly identified disease as nodular or unifocal disease in 80%, 92%, and 72% of the patients, respectively. Multicentric or diffuse disease was correctly identified by non-blinded PET, blinded PET, and CT in 44%, 31%, and 69%. Hence, non-blinded and blinded PET correctly identified the gross pathologic features of tumors in 27 (66%) and 28 (68%) of the patients respectively, resulting in a high concordance rate of 90% between the blinded and nonblinded PET interpretations. Gross tumor features identified by CT are listed in Table 2. Accuracy of PET and CT in identifying lymph node involvement The lymph node status of 35 of the 41 patients (85%) was assessed surgically or preoperatively by bronchoscopy and mediastinoscopy. Six patients (15%) were not evaluated for lymph node stage. Twenty-eight patients (68%) had no pathologic evidence for lymph node involvement. Of these, blinded and non-blinded PET correctly identified 25 (89%) and CT identified 22 (79%). PET and CT were false-positive in three and six patients respectively. Among the seven patients (17%) with lymph node involvement, both blinded and non-blinded PET as well as CT correctly identified positive lymph nodes in only two patients (29%). The average size of the positive lymph nodes was 1.57±0.64 cm, with a size range of 1–2.5 cm. PET tended to be false negative for perihilar lymph nodes that approximated 1 cm in size. CT was false negative in three patients with perihilar lymph nodes and in two with lymph nodes that were greater than 2.2 cm. The overall accuracy of non-blinded PET, blinded PET, and CT for lymph node staging was 77%, 77%, and 69% respectively (P=NS).
Comparison of histologic and metabolic parameters Statistical analysis of pathologic and metabolic parameters yielded significant correlations between visual FDG uptake and tumor size (r2=0.77, P=0.04) and BAC component (r2=–0.55, P=0.01). The SUV was unrelated to histologic grade and BAC component. Discussion This study demonstrates that FDG-PET detects BAC with a similar sensitivity as those published for the detection of other types of non-small cell lung carcinoma [1, 28]. This high sensitivity was achieved by both blinded and non-blinded PET image analysis. However, not all pure BAC lesions (100% BAC component) were detected. FDG-PET is also limited in its ability to detect mucinous tumors and small lesions (<1 cm). Finally, PET more accurately detects nodular and poorly differentiated than multicentric and well-differentiated BAC. The current results of an overall high sensitivity of PET for BAC detection contradict previous studies, which reported frequent false-negative PET findings in BAC [16, 17, 18, 19, 20]. Importantly, the previous studies also included tumors with varying components of BAC. The small number of patients enrolled in these previous studies might account for this discrepancy. The current results demonstrate that FDG-PET is as sensitive (89%) for detecting tumors with BAC component as it is for detecting other non-small cell lung cancer types [1, 28] (Fig. 1). However, the sensitivity of PET for detecting pure BAC according to the new WHO classification is low. PET missed four of the six pure BAC lesions, resulting in an overall sensitivity of 33%. This finding is more concordant with the findings of false negative PET studies in BAC [16, 17, 18, 19, 20], although it is unknown how many pure BACs were included in these previous reports. The sensitivities of the blinded and non-blinded PET interpretations were highly concordant for disease detection and identification of tumor gross features. The addition of attenuation-corrected images in the subgroup of 12 patients did not influence the classification of tumor gross features (multifocal vs unifocal), confirming reports that attenuation correction might not result in an improved diagnostic accuracy of PET [24, 29, 30]. CT was less accurate than PET in identifying nodular disease but more accurate in identifying multicentric disease. This is likely explained by the superior spatial resolution of CT. On the other hand, the higher sensitivity of PET comes at the expense of a lower specificity. PET tended to classify multicentric disease as nodular, which characteristically manifests as multiple small tumor volumes or as dispersed clusters of micronodules which are well below the spatial resolution of PET [31, 32].
European Journal of Nuclear Medicine Vol. 29, No. 9, September 2002
1171
Fig. 1. Coronal FDG-PET image of a 69-year-old female patient showing focally increased FDG uptake (SUV=1.8) in the left upper lobe of the lung (arrow). This lesion was later resected and was verified by pathology to be a non-mucinous, moderately differentiated pure BAC (BAC5)
This study also investigated the relationship between pathologic characteristics of BAC and PET findings. To this end, the histology of each tumor was reclassified according to the newer, more stringent 1999 WHO guidelines for diagnosing and defining pure BAC. While all 46 tumors of the 41 patients in our study population are BACs according to the previous, less stringent 1986 WHO definition, only six (13%) of these are pure BACs by the current definition. In addition to histologic tumor characteristics, the small size of some BAC lesions also contributed to falsenegative PET studies, since more than half of the missed lesions were less than 8 mm in size. The notion that lesion size is important for tumor detection with PET is
also supported by the significant correlation between FDG uptake and tumor size (P=0.04). Another probable source of false negative PET is the mucinous nature of some BACs, as previously reported [17, 33, 34]. In our study, two of four mucinous lesions were missed by PET. The presence of mucin is associated with low tumor cellularity [34] (Fig. 2). Mucinous carcinomas are commonly found in the gastrointestinal tract and studies in colorectal cancers have reported a lower accuracy of FDG-PET for tumors containing more than 50% mucin [35, 36, 37, 38]. Sensitivities of PET and CT did not differ significantly in identifying absence (89% and 79%, respectively) or presence (29% for both) of lymph node involvement. Sources of false negative lymph nodes on PET included small size or close proximity of lymph nodes to tumors. Our study found a good correlation between visual FDG uptake and BAC component (P=0.01). While there was little correlation with histologic grade, PET appeared to be especially sensitive for detection of poorly differentiated tumors. The average SUV value of all BACs in our study was 3.0±1.4. This is consistent with values previously reported by Kim et al. [17], who found lower SUVs in BAC (3.5) than in non-BAC adenocarcinoma (8.8) and squamous cell carcinoma (10.8). This study has several limitations. The number of pure BACs (BAC5; n=6) was low. This is because pure BAC is a rare entity. Secondly, semiquantitative FDG uptake data were only obtained in ten patients, thus greatly limiting the comparisons of SUV with histologic parameters. Of the six patients with pure BAC, only three had attenuation-corrected images. Attenuation correction did not add additional information in these three patients. However, additional diagnostic information obtained from corrected images cannot be ruled out in the three remaining patients. It should be noted that the additional value of semiquantitative analysis of tumor tracer uptake using the SUV has been questioned [39]. As another limitation, only patients who underwent surgical resection of the tumor after the PET study were included. Thus, these patients did not have advanced, non-surgically resectable disease. Therefore, the current findings can only be applied to patients without advanced BAC. Lastly, this retrospective evaluation was performed on patients with pathologically verified BAC. Specificity and overall accuracy can therefore not be determined from the current data. In summary, the diagnostic performance of wholebody FDG-PET is similar in patients with adenocarcinoma with a BAC component and in those with other nonsmall cell lung cancer types. However, metabolic imaging with FDG-PET might miss some of the pure BAC lesions.
European Journal of Nuclear Medicine Vol. 29, No. 9, September 2002
1172
Fig. 2. A Axial CT scan of a 70-year-old female patient showing two subpleural nodules (arrowheads) in the right lower lobe of the lung. B Transaxial PET scan of the same patient showing only one subpleural nodule. Both nodules (see A) were resected and the larger lesion (filled arrowhead) detected by both CT (A) and PET (B) was verified by pathology to be a 1.5-cm, non-mucinous, moderately differentiated tumor with 1%–24% BAC features (BAC1). The smaller lesion (unfilled arrowhead) missed by PET was verified to be a 0.8-cm, mucinous, well-differentiated pure BAC (BAC5). C Histology of a non-mucinous pure BAC (BAC5) at high magnification. Note the neoplastic cells growing along the alveolar walls (arrows). D Histology of a mucinous pure BAC (BAC5). Arrows indicate areas (pink) of mucin production
References 1. Gould MK, Maclean CC, Kuschner WG, Rydzak CE, Owens DK. Accuracy of positron emission tomography for diagnosis of pulmonary nodules and mass lesions: A metaanalysis. JAMA 2001; 285:914–924. 2. Vansteenkiste JF, Stroobants SG, DeLeyn PR, et al. Lymph node staging in non-small-cell lung cancer with FDG-PET scan: a prospective study on 690 lymph node stations from 68 patients. J Clin Oncol 1998; 16:2142–2149.
3. Gambhir SS, Hoh CK, Phelps ME, Madar I, Maddahi J. Decision tree sensitivity analysis for cost-effectiveness of FDGPET in the staging and management of non-small-cell lung carcinoma. J Nucl Med 1996; 37:1428–1436. 4. Brown RS, Leung JY, Kison PV, Zasadny KR, Flint A, Wahl RL. Glucose transporters and FDG uptake in untreated primary human non-small cell lung cancer. J Nucl Med 1999; 40:556–565. 5. Higashi K, Ueda Y, Sakurai A, et al. Correlation of Glut-1 glucose transporter expression with [F-18]FDG uptake in nonsmall cell lung cancer. Eur J Nucl Med 2000; 27:1778– 1785. 6. Higashi K, Ueda Y, Ayabe K, et al. FDG PET in the evaluation of the aggressiveness of pulmonary adenocarcinoma: correlation with histopathological features. Nucl Med Commun 2000; 21:707–714. 7. Leibow AA. Bronchiolo-alveolar carcinoma. Adv Intern Med 1960; 10:329–358. 8. Berkmen YM. The many faces of bronchiolo-alveolar carcinoma. Semin Roentgenol 1977; 12:207–214. 9. Epstein DM. Bronchioloalveolar carcinoma. Semin Roentgenol 1990; 25:105–111. 10. Hill CA. Bronchioloalveolar carcinoma: a review. Radiology 1984; 150:15–20. 11. Adler B, Padley S, Miller RR, Muller NL. High-resolution CT of bronchioloalveolar carcinoma. AJR 1992; 159:275–277.
European Journal of Nuclear Medicine Vol. 29, No. 9, September 2002
1173 12. Barsky SH, Cameron R, Osann KE, Tomita D, Holmes EC. Rising incidence of bronchioloalveolar lung carcinoma and its unique clinicopathologic features. Cancer 1994; 73:1163– 1170. 13. Barkley JE, Green MR. Bronchioloalveolar carcinoma. J Clin Oncol 1996; 14:2377–2386. 14. Travis WD, Sobin LH. Histological typing of lung and pleural tumours, 3rd edn. WHO, International Histological Classification of Tumours. Berlin Heidelberg New York: Springer, 1999. 15. World Health Organization. Histological typing of lung tumours, 2nd edn. WHO, International Histological Classification of Tumours. Geneva: World Health Organization, 1981. 16. Higashi K, Ueda Y, Seki H, et al. Fluorine-18-FDG PET imaging is negative in bronchioloalveolar lung carcinoma. J Nucl Med 1998; 39:1016–1020. 17. Kim BT, Kim Y, Lee KS, et al. Localized form of bronchioloalveolar carcinoma: FDG PET findings. AJR 1998; 170:935– 939. 18. Jang HJ, Lee KS, Kwon OJ, Rhee CH, Shim YM, Han J. Bronchioloalveolar carcinoma: focal area of ground-glass attenuation at thin-section CT as an early sign. Radiology 1996; 199:485–488. 19. Scott WJ, Schwabe JL, Gupta NC, et al. Positron emission tomography of lung tumors and mediastinal lymph nodes using [F-18]fluorodeoxyglucose. Ann Thorac Surg 1994; 58:698– 703. 20. Smith GT, Hubner KF, Peterson A, Hunter K, Neff J. FDG PET for evaluation of bronchioloalveolar cell carcinoma (BAC) of the lung. Clin Positron Imaging 1998; 1:260. 21. Minn H, Leskinenkallio S, Lindholm P, et al. [F-18]Fluorodeoxyglucose uptake in tumors: kinetic vs steady-state methods with reference to plasma insulin. J Comput Assist Tomogr 1993; 17:115–123. 22. Dahlbom M, Hoffman EJ, Hoh CK, et al. Whole-body positron emission tomography. Part I. Methods and performance characteristics. J Nucl Med 1992; 33:1191–1199. 23. Adam LE, Zaers J, Ostertag H, Trojan H, Bellemann ME, Brix G. Performance evaluation of the whole-body PET scanner ECAT EXACT HR+ following the IEC standard. IEEE Trans Nucl Sci 1997; 44:1172–1179. 24. Bengel FM, Ziegler SI, Avril N, Weber W, Laubenbacher C, Schwaiger M. Whole-body positron emission tomography in clinical oncology: comparison between attenuation-corrected and uncorrected images. Eur J Nucl Med 1997; 24:1091–1098. 25. Meikle SR, Dahlbom M, Cherry SR. Attenuation correction using count-limited transmission data in positron emission tomography. J Nucl Med 1993; 34:143–144.
26. Meikle SR, Hutton BF, Bailey DL, Hooper PK, Fulham MJ. Accelerated EM reconstruction in total-body PET: potential for improving tumour detectability. Phys Med Biol 1994; 39:1689–1704. 27. Mountain CF. Revisions in the International System for Staging Lung Cancer. Chest 1997; 111:1710–1717. 28. Haberkorn U. Positron emission tomography (PET) of nonsmall cell lung cancer. Lung Cancer 2001; 34:S115–S121. 29. Yasuda S, Ide M, Takagi S, et al. Cancer detection with wholebody FDG PET images without attenuation correction. Kaku Igaku 1996; 33:367–373. 30. Kotzerke J, Guhlmann A, Moog F, Frickhofen N, Reske SN. Role of attenuation correction for fluorine-18 fluorodeoxyglucose positron emission tomography in the primary staging of malignant lymphoma. Eur J Nucl Med 1999; 26:31–38. 31. Kubota K, Matsuzawa T, Fujiwara T, et al. Differential diagnosis of lung tumor with positron emission tomography: a prospective study. J Nucl Med 1990; 31:1927–1932. 32. Dewan NA, Gupta NC, Redepenning LS, Phalen JJ, Frick MP. Diagnostic efficacy of PET-FDG imaging in solitary pulmonary nodules: potential role in evaluation and management. Chest 1993; 104:997–1002. 33. Higashi K, Nishikawa T, Seki H, et al. Comparison of fluorine-18-FDG PET and thallium-201 SPECT in evaluation of lung cancer. J Nucl Med 1998; 39:9–15. 34. Berger KL, Nicholson SA, Dehdashti F, Siegel BA. FDG PET evaluation of mucinous neoplasms: correlation of FDG uptake with histopathologic features. AJR 2000; 174:1005–1008. 35. Whiteford MH, Whiteford HM, Yee LF, et al. Usefulness of FDG-PET scan in the assessment of suspected metastatic or recurrent adenocarcinoma of the colon and rectum. Dis Colon Rectum 2000; 43:759–767. 36. Flamen P, Stroobants S, Van Cutsem E, et al. Additional value of whole-body positron emission tomography with fluorine18-2-fluoro-2-deoxy-D-glucose in recurrent colorectal cancer. J Clin Oncol 1999; 17:894–901. 37. Valk PE, Abella-Columna E, Haseman MK, et al. Whole-body PET imaging with [F-18]fluorodeoxyglucose in management of recurrent colorectal cancer. Arch Surg 1999; 134:503–511. 38. Delbeke D, Vitola JV, Sandler MP, et al. Staging recurrent metastatic colorectal carcinoma with PET. J Nucl Med 1997; 38:1196–1201. 39. Valk PE, Pounds TR, Hopkins DM, et al. Staging non-small cell lung cancer by whole-body positron emission tomographic imaging. Ann Thorac Surg 1995; 60:1573–1581.
European Journal of Nuclear Medicine Vol. 29, No. 9, September 2002