Interact CardioVasc Thorac Surg 2007;6:350-353. doi:10.1510/icvts.2006.150664
© 2007 European Association of Cardio-Thoracic Surgery
Institutional report - Thoracic general |
Factors causing inaccurate staging of mediastinal nodal involvement in non-small cell lung cancer patients staged by positron emission tomography
Nael Al-Sarrafa,*,
Rashid Aziza,
Kishore Doddakulaa,
Kathy Gatelya,
Lorraine Wilsonb,
Eillish McGoverna and
Vincent Younga
a Department of Cardiothoracic Surgery, St James's Hospital, Dublin 8, Ireland
b Department of Nuclear Medicine, Blackrock Clinic, Dublin, Ireland
Received 19 December 2006;
received in revised form 22 February 2007;
accepted 6 March 2007
*Corresponding author. Tel.: +353-1-4103389; fax: +353-1-4103000.
E-mail address: trinityq8{at}hotmail.com (N. Al-Sarraf).
 |
Abstract
|
|---|
Despite documented superiority of positron emission tomography over other investigative modalities in the preoperative staging of non-small cell lung cancer, a proportion of patients will have an inaccurate staging of their mediastinal nodes. The aim of this retrospective review is to analyse the clinicopathological factors responsible for inaccurate nodal staging by integrated PET-CT. A total of 100 consecutive patients with histologically proven non-small cell lung cancer underwent staging with PET-CT prior to lung resection. Thirty-three patients, inaccurately staged by PET-CT, were analysed. Univariate analysis identified the following as significant in causing inaccurate nodal staging: history of tuberculosis (P=0.039) and non-insulin dependant diabetes (P=0.014). In multivariate analysis, we have identified the following as independent factors in causing inaccurate staging of mediastinal lymph nodes: rheumatoid arthritis, non-insulin dependent diabetes, history of tuberculosis, presence of atypical adenomatous hyperplasia and pneumonia (P<0.05). The highest rate of inaccuracy in mediastinal nodal staging was in nodal station 4 (11%, P=0.01) followed by station 7 (10%, P=0.02) and station 9 (3.5%, P=0.01). Interpretation of PET-CT staging of the mediastinum in patients with a history of the above should be with caution, as the incidence of false upstaging and down staging in these subgroups is high. Vigilance of such factors may improve the accuracy of PET-CT in staging mediastinal lymph nodes. Histological confirmation should always be sought.
Key Words: PET/CT; Non-small cell lung cancer; Inaccurate staging
|
1. Introduction
|
|---|
PET-CT has emerged as an accurate non-invasive staging modality in detecting lymph node involvement in non-small cell lung cancer (NSCLC). Despite its high sensitivity, specificity and accuracy, there are still a proportion of patients who are falsely upstaged or downstaged by PET-CT. PET detection of primary NSCLC and its associated lymph nodal involvement is solely based on the metabolic uptake of 2-[18F] fluoro-2-deoxy-D-glucose (FDG) by tumours. This tracer is taken up by the tumour through a carrier mediated transporter and subsequently becomes phosphorylated by hexokinases which leads to entrapment in the tumour tissue without significant uptake into normal healthy tissue [1]. However, in certain circumstances, there will be some uptake into active normal tissues or tissues involved by inflammatory processes [2]. This could lead to inaccurate staging and, thus, potential mis-management unless the uptake is verified by invasive staging tools (e.g. mediastinoscopy).
In this retrospective study, we analysed a series of clinicopathological factors that may be responsible for falsely upstaging and downstaging tumours in relation to lymph node involvement. We believe that the staging accuracy of PET-CT could be improved by the identification of factors that may interfere with correct staging and, thus, correct management.
|
2. Materials and methods
|
|---|
2.1. Patients
Between September 2004 and January 2006, 100 consecutive patients with histologically proven NSCLC underwent staging by PET-CT as an adjunct to CT prior to lung resection. All clinical, operative, radiological and pathological findings were analysed retrospectively. Histological classification of NSCLC is based on that of WHO classification [3]. Pre and postoperative staging were based on the tumour-node-metastasis (TNM) staging system [4]. No patient had pre-operative chemo radiotherapy. Ethical approval for this study was obtained from the hospital ethical committee.
2.2. Imaging
All patients had CT and integrated PET-CT images performed within four weeks of surgery. PET-CT was performed using integrated PET-CT scanner (Discovery ST, GE Medical systems). Patients were fasting for 6 h prior to the procedure and the images were obtained after 60 min of intravenous administration of 18-fluoro-deoxy-glucose (18-FDG). All patients were asked to remain resting and to void just before scanning. Attenuation corrected and non-attenuation corrected images were obtained and read by an experienced nuclear physician. Lesions were considered pathological (i.e. positive) if a definite localized area of higher FDG uptake than the surrounding normal tissue exist (excluding physiological uptake). All positive nodes were localised according to the American Thoracic Society (ATS) guidelines [5]. False upstaging by PET-CT is defined as incorrectly staging N0 stage to N1/N2 or N1 to N2 (when compared to the pathological staging). False downstaging is defined as incorrectly staging lymph nodes from N1/N2 to N0 or staging N2 to N1 (when compared to pathological staging). Inaccurate (or incorrect) staging by PET is defined as combined false upstaging and false downstaging (when compared to pathological staging). Standardised uptake values were calculated using an in-built software programme that takes into account body weight, dose of tracer and activity at a pixel defined by a region of interest. Regions of interest (ROI) were manually drawn on the trans-axial images around focal FDG-uptake zones of the primary malignancy and lymph nodes. To minimise partial volume effects, we used maximum SUV within a region of interest (SUVmax).
2.3. Surgical resection
A total of 89 patients had lung resections and systematic lymph node dissections by two experienced cardiothoracic surgeons (EM and VY). Of the remaining patients, nine had only cervical mediastinoscopy/anterior mediastinotomy and were considered histologically unresectable (diffuse N2 disease) and two had open/close thoracotomies due to widespread N2 disease intra-operatively confirmed on frozen sections. All lymph nodes were examined by a pulmonary histopathologist for the presence/absence of malignancy. All resected lung specimens were evaluated for presence/absence of pulmonary disease. Systemic lymph node dissection was employed in all cases that had resection. Nodal stations 2,4,7,8,9 and 10 were resected in right-sided lung resections while nodal stations 5,6,7,8,9 and 10 were resected in left-sided resections. Bronchopulmonary nodes (N1) are removed as part of the lung resection specimen. Cervical mediastinoscopy was used to biopsy nodes in stations 2,4 and superior part of 7. Anterior mediastinotomy was performed for nodal stations 5 and 6.
2.4. Data analysis
Findings on PET-CT images were compared with that of histopathological examination of the resected lymph node stations in order to determine their diagnostic sensitivity [TP/(TP+FN)], specificity [TN/(TN+FP)] and accuracy [(TP+TN)/n] with TP=true positive, TN=true negative, FP=false positive, FN=false negative, n=total number of patients. The software package SPSS v11.0 (SPSS Inc., Chicago, IL) was used to perform the statistical analysis. Categorical data were compared using the Fisher's exact test. Multivariate analysis was carried out using logistic regression (backwards stepwise) method. P-values were considered statistically significant if <0.05.
|
3. Results
|
|---|
Our cohort was comprised of 53 men and 47 women with an age range of 25–81 years and mean (±S.D.) of 63 (±10). All patients enrolled had histologically proven NSCLC. There were equal numbers of adenocarcinomas (AC) and squamous cell carcinomas (SQCC) (50 patients each). Primary tumour size, as determined by measuring the maximum diameter on CT, has ranged overall from 1 cm to 8.5 cm with a mean (±S.D.) of 3.4 cm (±1.6 cm). A total of 609 nodal stations were studied, averaging six nodal stations per patient. With regard to nodal staging, PET-CT has correctly staged 67 patients (67%) but falsely downstaged 25 patients (25%) and falsely upstaged eight patients (8%). Accuracy of PET-CT in mediastinal node staging was 67% with a sensitivity of 48% and specificity of 93%.
Clinicopathological factors were studied and their distribution among the accurately and inaccurately staged groups are summarised in Table 1. All patients with rheumatoid arthritis were in the inactive stage of their illness at the time of PET scanning. Diabetics had optimum control of their diabetes prior to scanning with their BM levels <8.0 mmol/l. Univariate and multivariate analyses are summarised in Tables 2 and 3, respectively. A second univariate analysis was performed to assess the accuracy of PET-CT in nodal staging (as compared to pathological evaluation) on a station-by-station basis and is summarised in Table 4.
|
4. Discussion
|
|---|
Integrated PET-CT staging of nodal disease remains the best of all the non-invasive staging modalities with many studies showing higher accuracy than that of conventional CT [6, 7]. Our aim in this study was to assess factors responsible for false staging (i.e. upstaging and downstaging). Clinical factors including: rheumatoid arthritis, type 2 diabetes mellitus (DM) and history of TB can lead to false staging of the tumours. In the case of rheumatoid arthritis, none of our patients were in the active stage of the disease. However, there was a 2.1-fold increase in inaccurate staging (21% vs. 10%, P=0.048). This can be explained by the fact that inflammatory conditions have increased glucose uptake by inflammatory cells and macrophages, and this in turn can cause false emission of the tracer which will be detected by PET [8]. Patients with a history of TB, although none of them were in the active or in the treatment phase, had an eight-fold increase in their inaccurate staging (12% vs. 1.5%, P=0.038). These results are consistent with earlier reports from Knoishi et al. [9]. In his study, the authors have shown that half of their overstaged patients had evidence of previous exposure to TB with all falsely positive nodes showing evidence of lymph node hyperplasia histologically [9]. DM is another factor causing inaccurate staging with an increased risk of false staging (15% vs. 1.5%, P=0.017). All our patients with DM were type 2 and had well-controlled diabetes at the time of scanning as reflected by their measured BMs prior to injecting the tracer. This finding is consistent with other studies which have demonstrated that the accuracy of FDG-PET is dependant on serum glucose levels and that the false negative results of PET are mostly encountered in patients with raised serum glucose [10]. A number of investigators have mentioned that blood sugar competes with FDG in the glucose uptake by tumours and that hyperglycaemia impairs the FDG uptake in tumours [11]. These findings were reported in hyperglycaemic patients as opposed to those with well controlled BM level.
There were statistically significant differences in the pathological factors that may contribute to false staging of lymph node involvement (whether upstaging or downstaging). Such factors include: pneumonic changes histologically and the presence of atypical adenomatous hyperplasia (Table 3). Evidence of pneumonia in the resected lung tissue was associated with higher false staging than accurate staging (36% vs. 19%, P=0.012). This is consistent with previous reports from Takamochi et al. [12]. In their study, pneumonia was a factor in the false positive staging of NSCLC. This may well be explained by active uptake of glucose into these tissues causing inaccurate results [12]. Finally, AAH (defined as a focal lesion in which the involved alveoli and respiratory bronchi are lined by monotonus, slightly atypical cuboidal to low columnar epithelial cells with dense nuclear chromatin) histologically seems to be associated with a 3-fold increase in false staging when compared to accurately staged group (12% vs. 4%, P=0.041). This is consistent with the pathophysiology, in which AAH arises as a result of chronic inflammation and is considered by some as a pre-neoplastic event [13].
We have failed to show any statistical significance in other factors studied such as site, size of primary tumours, NSCLC subtype, grade of pathological differentiation, standardised uptake values of primary NSCLC, the presence of lung scarring and emphysematous changes histologically in the resected lung tissues (Tables 2 and 3). Clinical diagnosis of COPD showed no statistical significance in relation to inaccurate nodal staging by PET (P=0.94). It should be emphasised that all our patients with COPD were not in the active (infective exacerbation) stage at the time of staging by PET and their FDG uptake may not become detectable to a level that can cause statistical significance. We also have to emphasise that although our cohort was comprised of an equal number of AC and SQCC, the incidence of inaccurate staging was higher in SQCC vs. AC (55% vs. 39%). However, this has not reached statistical significance. It has been reported that SQCC has stronger Glut-1 expression that that of AC [14] and since FDG uptake is known to be closely related to over-expression of Glut-1 in human cancers [15], it seems that this over-expression could enhance the uptake of glucose into mediastinal lymph nodes and potentially leading to false positive results. Our results confirm that reported by Knoishi et al. [9] which showed no relation between false positive nodes and histological typing in his study.
In our study, moderately differentiated tumours had higher rate of inaccurate nodal staging when compared to that of poor differentiated tumours (70% vs. 30%, Table 2), this has not reached statistical significance. This may be merely a reflection of the differential uptake of FDG by the various pathological grades of tumour differentiation and their corresponding proliferation rates. It has to be emphasised, however, that some association may well exist, but larger numbers of patients need to be analysed to show the significance of such an association. We had a higher incidence of lung scarring (idiopathic fibrosis) histologically in the inaccurately staged group when compared to those with accurate nodal staging (12% vs. 7.5%, Table 2). However, such an association was not statistically significant and, as such, further studies are warranted to determine the possibility of such an association.
Finally, the highest rate of inaccuracy among mediastinal nodal stations was in ATS station 4, followed by ATS station 7 and ATS 9 (Table 4). This is related to the close proximity of these nodes to hilar nodes (N1) and PET remains limited in its ability to reliably distinguish N2 from N1 nodes. It becomes imperative that these stations should be verified histologically when FDG uptake is high in order to determine the preoperative stage and the appropriate treatment strategy (i.e. surgery vs. chemotherapy).
In conclusion, we have identified some clinicopathological factors that can cause inaccurate lymph node staging by integrated PET-CT. To improve the accuracy of PET-CT, one must be cautious when interpreting PET-CT results in the presence of these identified factors. Histological confirmation should be sought when appropriate.
|
References
|
|---|
- Wahl RL, Hutchins G, Buchsbaum D, Liebert M, Grossman HB, Fisher S. 18-F-2-deoxy-2-fluoro-D-glucose (FDG) uptake in human tumour xenografts: feasibility studies for cancer imaging with PET. Cancer 1991; 67:1544–1550.[CrossRef][Medline]
- Bakheet SM, Powe J. Benign causes of 18-FDG uptake on whole body imaging. Semin Nucl Med 1998; 28:352–358.[CrossRef][Medline]
- World Health Organization, histological typing of lung and pleural tumours. 3rd edition 1999;New York: Springer-Verlag.
- Mountain CF. Revisions in the international system for staging lung cancer. Chest 1997; 111:1710–1717.[CrossRef][Medline]
- Tisi GM, Friedman PH, Peters RM. American Thoracic Society: clinical staging of primary lung cancer. Am Rev Respir Dis 1983; 127:659–664.[Medline]
- Antoch G, Stattaus J, Nemat AT, Marnitz S, Beyer T, Kuehl H, Bock A, Debatin JF, Freudenberg LS. Non-small cell lung cancer: dual-modality PET/CT in preoperative staging. Radiology 2003; 229:526–533.[Abstract/Free Full Text]
- Van Tinteren H, Hoekstra O, Smit E, Van den Bergh JH, Schreurs AJ, Stallaert RA, van Velthoven PC, Comans EF, Diepenhorst FW, Verboom P, van Mourik JC, Postmus PE, Boers M, Teule GJ. Effectiveness of PET in the preoperative assessment of patients with suspected non-small cell lung cancer: the PLUS multicenter randomised trial. Lancet 2002; 359:1388–1392.[CrossRef][Medline]
- Kubota R, Kubota K, Yamada S, Tada M, Ido T, Tamaahashi N. Microautoradiographic study for the differentiation of intratumoral macrophages, granulation tissues and cancer cells by the dynamics of fluorine-18-Fluorodeoxyglucose uptake. J Nucl Med 1994; 35:104–112.[Abstract/Free Full Text]
- Konishi J, Yamazaki K, Tsukamoto E, Tamaki N, Onodera Y, Otake T, Morikawa T, Kinoshita I, Dosaka-Akita H, Nishimura M. Mediastinal lymph node staging by FDG-PET in patients with non-small cell lung cancer: analysis of false-positive FDG-PET findings. Respiration 2003; 70:500–506.[CrossRef][Medline]
- Zimny M, Bares R, Fass J, Adam G, Cremerius U, Dohmen B, Klever P, Sabri O, Schumpelick V, Buell U. Fluorine-18-fluorodeoxyglucose positron emission tomography in the differential diagnosis of pancreatic carcinoma: a report of 106 cases. Eur J Nucl Med 1997; 24:678–682.[Medline]
- Gorenberg M, Hallett WA, O'Doherty MJ. Does Diabetes affect [18F] FDG standardized uptake values in lung cancer? Eur J Nucl Med Mol Imaging 2002; 29:1324–1327.[CrossRef][Medline]
- Takamochi K, Yoshida J, Murakami K, Niho S, Ishii G, Nishimura M, Nishiwaki Y, Suzuki K, Nagai K. Pitfalls in lymph node staging with positron emission tomography in non-small cell lung cancer patients. Lung Cancer 2005; 47:235–242.[CrossRef][Medline]
- Kayser K, Nwoye JO, Kosjerina Z, Goldmann T, Vollmer E, Kaltner H, Andre S, Gabius HJ. Atypical Adenomatous hyperplasia of lung: its incidence and analysis of clinical, glycohistochemical and structural features including newly defined growth regulators and vascularization. Lung Cancer 2003; 42:171–182.[CrossRef][Medline]
- Chung JH, Cho KJ, Lee SS, Baek HJ, Park JH, Cheon GJ, Choi CW, Lim SM. Overexpression of Glut-1 in lymphoid follicles correlates with false-positive 18F-FDG PET results in lung cancer staging. J Nucl Med 2004; 45:999–1003.[Abstract/Free Full Text]
- Higashi T, Ueda Y, Sakurai A, Wang XM, Xu L, Murakami M, Seki H, Oguchi M, Taki S, Nambu Y, Tonami H, Katsuda S, Yamamoto I. Correlation of Glut-1 glucose transporter expression with [18F] FDG uptake in non-small cell lung cancer. Eur J Nucl Med 2000; 27:1778–1785.[CrossRef][Medline]
This article has been cited by other articles:
|
|
|
|
 
A. Bille, E. Pelosi, A. Skanjeti, V. Arena, L. Errico, P. Borasio, M. Mancini, and F. Ardissone
Preoperative intrathoracic lymph node staging in patients with non-small-cell lung cancer: accuracy of integrated positron emission tomography and computed tomography
Eur. J. Cardiothorac. Surg.,
September 1, 2009;
36(3):
440 - 445.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
M. Morikawa, Y. Demura, T. Ishizaki, S. Ameshima, I. Miyamori, M. Sasaki, T. Tsuchida, H. Kimura, Y. Fujibayashi, and H. Okazawa
The Effectiveness of 18F-FDG PET/CT Combined with STIR MRI for Diagnosing Nodal Involvement in the Thorax
J. Nucl. Med.,
January 1, 2009;
50(1):
81 - 87.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
N. Al-Sarraf, K. Gately, J. Lucey, R. Aziz, K. Doddakula, L. Wilson, E. McGovern, and V. Young
Clinical implication and prognostic significance of standardised uptake value of primary non-small cell lung cancer on positron emission tomography: analysis of 176 cases
Eur. J. Cardiothorac. Surg.,
October 1, 2008;
34(4):
892 - 897.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
D. Uesaka, Y. Demura, T. Ishizaki, S. Ameshima, I. Miyamori, M. Sasaki, Y. Fujibayashi, and H. Okazawa
Evaluation of Dual-Time-Point 18F-FDG PET for Staging in Patients with Lung Cancer
J. Nucl. Med.,
October 1, 2008;
49(10):
1606 - 1612.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
N. Al-Sarraf, R. Aziz, K. Gately, J. Lucey, L. Wilson, E. McGovern, and V. Young
Pattern and predictors of occult mediastinal lymph node involvement in non-small cell lung cancer patients with negative mediastinal uptake on positron emission tomography
Eur. J. Cardiothorac. Surg.,
January 1, 2008;
33(1):
104 - 109.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
Top
Abstract
1. Introduction