Abstract

Background: This study aims to determine a new cut-off value for standardized uptake value in positron emission tomography-computed tomography evaluation of mediastinal lymph nodes in non-small cell lung cancer in Turkey.

Methods: A total of 207 patients with non-small cell lung cancer who were performed positron emission tomography-computed tomography between November 2006 and February 2010 were prospectively analyzed. Of these patients, 143 patients (125 males, 18 females; mean age 62.1 years; range 39 to 85 years) whose invasive staging was performed after positron emission tomography were included in the study. Sensitivity, specificity, positive predictive value, negative predictive value, and accuracy rates of positron emission tomography were calculated and compared using a maximum standardized uptake value cut-off value of ≥2.5 and the newly determined maximum standardized uptake value cut-off value.

Results: The sensitivity, specificity, positive predictive value, negative predictive value, and accuracy rates were 45.3%, 78.9%, 55.8%, 71%, and 66.4%, respectively, when the maximum standardized uptake value cut-off value was considered 2.5 in mediastinal lymph nodes. The new maximum standardized uptake value cut-off value was determined to be 4.8 in metastatic lymph nodes. These values were 39.6%, 91.1%, 72.4%, 71.9% and 72%, respectively, according to the new maximum standardized uptake value cut-off value of 4.8. There was a significant difference only between specificity rates when the two different maximum standardized uptake value cut-off values were used (p=0.022).

Conclusion: In this study, the sensitivity of positron emission tomography in the evaluation of mediastinal lymph nodes was lower than those reported in the literature. This situation may be associated with the frequently observed granulomatous infections such as tuberculosis in our country. Results of positron emission tomography should be evaluated according to countries and a new maximum standardized uptake value cut-off value should be calculated particularly for mediastinal lymph node metastasis in multicenter studies in our country.

Lung cancer ranks first among cancer-related deaths, accounting for 12% to 16% of all cancers and 17.8% to 28% of all cancer-related deaths. Despite advanced treatment applications, only 10% to 15% of the patients with lung cancer could survive more than five years after diagnosis.[1] For this reason, early diagnosis and establishment of new methods for the prevention of the disease are of utmost importance. Recently, positron emission tomography (PET) has become a very important imaging method in diagnosis, staging and monitoring of cancer patients, as well as for treatment planning. Since PET in clinical applications can display metabolic changes in cancer cells, it is an important auxiliary imaging technique in diagnosis of malignancy. Although there are many agents that can be used in PET imaging, glucose analogues targeting the increased glucose metabolism such as 18F-fluoro-deoxy-D-glucose (FDG) that are frequently radiolabeled in the images are used in many thoracic malignancies. Generally, a lesion with a standardized uptake value (SUV) of >2.5 is accepted as malign.[2]

Studies performed for non-small cell lung cancer (NSCLC), the most frequent cause of death due to malignancy, have revealed that mediastinal lymph node (MLN) metastasis is the most important prognostic factor while also the most important factor in defining the treatment for patients without distant metastasis.[3] Detection of MLN on thorax computed tomography (CT) provides anatomical information; however, thorax CT has limited sensitivity and specificity to detect the presence or absence of metastasis.[4] Computed tomography has a sensitivity of 57% and a specificity of approximately 82% in mediastinal staging. On the other hand, PET has higher accuracy, having sensitivity of 74.9% to 91%, specificity of 80% to 95%, negative predictive value (NPV) of 84% to 100%, and positive predictive value (PPV) of 60% to 93% in the assessment of MLN metastasis.[5]

In this study, we aimed to determine a new cutoff value for maximum standardized uptake value (SUV_max) in PET-CT evaluation of MLN in NSCLC in Turkey.

Methods

Two-hundred and seven patients who were examined by PET-CT due to the diagnosis of NSCLC between November 2006 and February 2010 in the Department of Thoracic Surgery of Medical Faculty of Ankara University were evaluated prospectively. Of these patients, 143 (125 males, 18 females; mean age 62.1 years; range 39 to 85 years) were included in the study whose lymph node sampling was performed by transbronchial needle aspiration (TBNA), mediastinoscopy, and thoracotomy. The study protocol was approved by the Medical Faculty of Ankara University Ethics Committee. Written informed consent was obtained from each patient, and the study was conducted in accordance with the principles of the Declaration of Helsinki.

All tests and invasive interventions were performed according to the 1997 tumor node metastasis (TNM) staging. Age, sex, fasting blood glucose level, and histories of diabetes mellitus, tuberculosis, chemotherapy and/or radiotherapy were evaluated. In addition, location of the mass, SUV_max of the mass and MLN, dimensions of MLN, cell type, and pathology results of the lymph node sampling were also analyzed.

Before PET-CT scan, patients were asked to fast for eight hours; thus, the blood glucose level of each patient was below 150 mg/dL (78-148 mg/dL) before the procedure. In each patient, images were obtained from the skull base to the upper thigh. After the measurements of the blood glucose level and blood pressure, 18F-FDG dose, which was calculated according to the body weight of the patients (0.15 mCi/kg), was administered intravenously for PET-CT scan. Following the injection, which was performed in the half-roll position in armchairs located in the waiting room, the images of the patients, who took a rest at an appropriate room temperature, were obtained after bladder emptying. For the thorax images, a PET-CT device (Discovery-STE 8, General Electric Medical System, Waukesha, WI, USA) that is a combination of a tomography unit with helical octagonal slice and a PET unit including bismuth germanate block detector was used.

Statistical analysis
Data were analyzed using the SPSS for Windows version 15.0 software (SPSS, Inc., Chicago, IL, USA). Categorical variables were expressed as number and percentages, whereas numerical variables were expressed as mean ± standard deviation. The differences between the groups were evaluated using the chi-square test, t test or one-way analysis of variance according to the types of the variables. Receiver operating characteristics (ROC) analysis was used to determine the most appropriate cut-off value for SUV_max level. In the comparison of diagnostic methods, the sensitivity, specificity, PPV, NPV, and accuracy measures were used. A p value of <0.05 was considered statistically significant.

Results

General characteristics of the patients are given in Table 1. According to the radiological evaluation of patients; TBNA, mediastinoscopy or surgical resection was performed. Of the patients, 132 (92.7%) underwent an operation. In the remaining non-operated 11 patients (8%), mediastinoscopy was performed on three patients and TBNA was performed on eight patients. In these patients, N2 was determined and treatment plan was changed from operation to adjuvant or neoadjuvant treatment. The preliminary case (6.9%) was operated after ruling out the lymph node metastasis by TBNA or mediastinoscopy in compliance with the preoperative findings. In preoperative staging, lymph node metastasis was observed in four patients after resection in pathologic examination of lymph node involvement. In these four patients, three had hilar lymph nodes adjacent to the mass on TBNA, and the right lower lobe tumors were paraesophageal lymph nodes in one patient. Location of the lymph nodes may have an effect on false negative. In these patients with false negative lymph nodes, the nodes were close to the mass.

Table 1: General characteristics of patients (n=143)

Lymph node metastasis was found in eight of the 14 patients with SUV_max o f > 2.5 and in 29 of the 92 patients with SUV_max of <2.5. Fifty patients (35%) had centrally located tumor and N₁ was determined in 21 (42%) of these. In 21 (42%) of the centrally located tumors, no involvement was determined in PET-CT; however, pathological examination revealed metastasis in 17 (40.5%). Non-central tumor was present in 93 patients (65%); while no involvement was determined in PET-CT in 12 (15.4%), metastasis was determined in hilar lymph node in pathological examination. The accuracy rate in determining hilar lymph node metastasis in PET-CT in centrally located tumors was significantly lower than in non-central tumors (p=0.005).

In the present study; sensitivity, specificity, PPV, NPV and accuracy of PET-CT were evaluated with SUV_max of 2.5 as the cut-off in N₁ and N₂ and a new SUV_max cut-off value was calculated by ROC analysis (Figure 1). In this analysis, the dependent variable with the SUV_max in the metastasis of lymph nodes and independent variable with the SUV_max in all lymph glands were used and the new cut-off value was calculated. The new SUV_max cut-off value was determined to be 5.7 for N₁ lymph nodes and 4.08 for N₂ lymph nodes. When N₂ and N₁ lymph nodes were evaluated together, the new SUV_max cut-off value was determined to be 4.8. The sensitivity, specificity, PPV, NPV, and accuracy of PET-CT in determining lymph node metastasis with SUV_max o f 5 .7, 4 .08, and 4.8, as well as comparison of these values with those calculated for SUV_max of 2.5 are presented in Tables 2 and 3.

Figure 1: Receiver operating characteristic curve for new maximum standardized uptake value cut-off value (4.8) in N₁ and N₂ lymph node stations.

Table 2: Sensitivity, specificity, positive and negative predictive values, and accuracy of positron emission tomography-computed tomography in determining N₁ with maximum standardized uptake value of 2.5, 5.7, and 4.08

Table 3: Sensitivity, specificity, positive and negative predictive values, and accuracy of positron emission tomography-computed tomography in determining N₁ and N₂ with maximum standardized uptake value of 2.5 and 4.8

The MLN metastasis rates in adenocarcinoma and squamous cell carcinoma were 41.7% and 36.4%, respectively. In the present study, cell type had no significant effect on MLN metastasis (p=0.686). Evaluation of the mass in terms of SUV_max and the possibility of lymph node metastasis revealed that the mean SUV_max of the mass with lymph node metastasis was 14.3±9.0, whereas the mean SUV_max of the mass without lymph node metastasis was 12.0±6.4, and the difference was not significant (p=0.08). Statistical evaluation of the SUV_max of the lymph node and the SUV_max of the primary mass in terms of risk for lymph node metastasis revealed that if the ratio of MLN SUV_max to the primary tumor SUV_max (lymph node SUV_max/primary mass SUV_max) was >0.29, the rate of metastatic possibility was determined to be high, and the sensitivity, specificity, PPV, NPV, and accuracy in this level were calculated to be 91%, 45%, 64.5%, 81.8%, and 69%, respectively.

Discussion

The use of TNM staging system is essential in malignant diseases in order to plan the treatment and predict the prognosis. In the treatment planning of patients with NSCLC, distinguishing patients suitable or unsuitable for surgical resection but who may benefit from chemotherapy and/or radiotherapy is of importance. Mediastinal lymph node involvement has been reported in 26% of newly diagnosed lung cancer patients, of whom extrathoracic metastasis has been determined in 49%.[3]

Studies on NSCLC, the most frequent cause of death due to malignancy, have demonstrated that MLN metastasis is the most important prognostic factor and also the most important factor in defining the treatment for patients without distant metastasis.[6] Tumor localization within the thorax, tumor size, tumor respectability, and its relationship with the surrounding anatomic structures can be determined using thorax CT. Presence of MLN can also be shown anatomically by thorax CT; however, it has limited sensitivity and specificity to show the presence or absence of metastasis. In MLN staging, the sensitivity and specificity of CT are approximately 57% and 82%, respectively.[4] Positron emission tomography-CT has become a very important imaging tool in planning the diagnosis, as well as in staging and monitoring of cancer patients. In the evaluation of MLN metastasis, the accuracy of PET-CT is higher than CT, with sensitivity of 74.9% to 91%, specificity of 80% to 95%, NPV of 84% to 100%, and PPV of 60% to 93%.[5]

Evaluation of NSCLC through non-invasive tests has demonstrated that the lymph nodes having the possibility to become metastatic should be verified by cytopathology. Transbronchial needle aspiration, endobronchial ultrasound-fine needle aspiration, endoscopic ultrasound, and transthoracic needle aspiration are newly developed minimally invasive techniques providing cytopathological diagnosis. While the specificity of these techniques is high, their NPV is low. These can be defined by minimal invasive surgical techniques (Table 4).[6] Mediastinoscopy is still the gold standard in the assessment of MLN.[7]

Table 4: Accuracy rate of different techniques in staging of mediastinal lymph nodes

In previous studies, while the specificity of PET-CT has been reported to be high in detecting metastatic lymph nodes, its sensitivity has been reported to be low.[6] In our study, when the mediastinal and hilar lymph nodes were evaluated together with SUV_max cut-off value as 2.5, the sensitivity and PPV were found to be lower as compared to those reported in other studies. In the study by Shin et al.[8] performed on 184 cases, the sensitivity and specificity of PETCT was reported to be 43% and 95%, respectively, for detecting N₂ disease in stage T₁ NSCLC. In the meta-analysis by Gould et al.[9] including 28 studies, the sensitivity was 61% and the specificity was 79%. In their study, Taşçı et al.[10] reported sensitivity, specificity, NPV, PPV, and accuracy of PET-CT in detecting N₂ disease to be 72%, 94.4%, 97.7%, 49.2%, and 92.7%, respectively. In our study, the sensitivity and PPV were determined to be low. In the study by Liu et al.,[11] the sensitivity and PPV were determined as 65% and 78.5%, respectively. Furthermore, in their study, Andrea et al.[12] investigated the accuracy rate of PET-CT in intrathoracic lymph node staging and determined the sensitivity, specificity, PPV, NPV and accuracy as 54.2%, 91.9%, 82.3%, 74.3%, and 80.5%, respectively. In the present study, the sensitivity was lower than and the specificity was similar to those reported in previous studies. Presence of inflammatory diseases is known to be the reason leading to false positive results on PET-CT and FDG uptake may be observed in the regions where active inflammation is present. It was considered that mediastinal involvement might be observed on PET-CT due to the presence of lower involvement, benign infections (tuberculosis, histoplasmosis) and inflammatory diseases. In the assessment of MLN by PET-CT, the cell diagnosis is required to prove the metastasis due to the lack of PPV. Before deciding on surgery, invasive/minimally invasive staging should be performed. However, since NPV was determined to be high, the treatment may be planned before staging if no involvement was determined in the MLN on PET-CT.

In the present study population, the SUV_max cut-off value was determined to be 4.8. With the use of this cut-off value, the sensitivity and PPV were determined to be low at a rate of 39.6% and 72.4%, respectively, which can lead to false negative results in the assessment of lymph nodes; i.e. metastasis. In their study, Lee et al.[13] determined a new cut-off value and suggested that despite its importance and efficacy in non-invasive staging, the efficacy of PET-CT was low due to the false-positive (infection, inflammatory diseases) and false negative (micrometastasis) results.

In regions where prevalence of granulomatous diseases is high, high FDG uptake may be monitored in benign lesions. In a study by Kwan et al.[14] conducted on healthy subjects in Taiwan, FDG uptake was monitored in the mediastina of 28% of the subjects. In their study, Kang et al.[15] determined a significant difference between the SUV_max of benign and malignant lesions.

In the present study, it was found that if the ratio of SUV_max of the MLN to the SUV_max of the primary tumor was >0.29, the rate of metastatic possibility was high at a rate of 91%; it was also detected that the lymph node could be malignant. In their study, Cerfolio et al.[16] i nvestigated t he p ossibility o f m etastasis according to the ratio of the SUV_max of the lymph node to that of the MLN on PET-CT, and determined in the pathologically proven cases that if the ratio of SUV_max of the primary tumor to that of MLN was >0.56, the lymph node could be malignant at a rate of 94%.

The SUV_max values were found to be significantly higher in the squamous cell carcinoma than in adenocarcinoma (p=0.02). Kim et al.[17] determined the SUV_max to be 10.8±4.4 for squamous cell carcinoma, and 8.8±3.2 for adenocarcinoma. The accuracy rate in determining hilar lymph node metastasis in PET-CT in centrally located tumors was significantly lower than in non-central tumors (p=0.005). It is difficult to detect the metastasis in lymph node that is adjacent to the primary tumor in PET-CT. N₁ lymph nodes are very close to primary tumors; thus, it may be overlooked because of FDG uptake of primary tumor. In the study by Al-Sarraf et al.,[18] the distinction of lymph node invasion was determined to be challenging in the tumors located in the hilum.

In their study, Lee et al.[19] investigated risk factors for occult mediastinal metastasis in NSCLC and showed that central location, larger tumor size, high SUV_max of PET-CT, and adenocarcinoma cell type were the risk factors. In the present study, in terms of lymph node metastasis, the N₁ risk was higher only in the centrally located tumors (p=0.005).

When the results of the present study were compared with those conducted in other developing countries, the accuracy rates were found to be lower. This might be due to the frequency of tuberculosis, fungal and other infections in Turkey. Positron emission tomography-CT results should be evaluated separately for each country and new SUV_max cut-off value should be determined considering our country’s realities in future multi-center studies for MLN, in particular.

Positron emission tomography-CT, a noninvasive staging method, has higher accuracy than other methods in the mediastinal lymph node staging; however, sufficient results have not been achieved. Although PET-CT is an appropriate technique, it is not as reliable as biopsy as the final diagnosis may be established by pathologic examination in the staging of mediastinal lymph node in lung cancer.

In conclusion, distinguishing patients suitable or unsuitable for surgical resection but who might benefit from chemotherapy and/or radiotherapy is important in non-small cell lung cancer. Recently, positron emission tomography-computed tomography is frequently used in lung cancer staging. However, decisions should be based on not only positron emission tomography-computed tomography findings but also on tumor location, cell type, and computed tomography findings interpreted by an experienced radiologist. As the rate of hilar lymph node metastasis is high particularly in the centrally located tumors, these tumors should be evaluated in detail.

Declaration of conflicting interests
The authors declared no conflicts of interest with respect to the authorship and/or publication of this article.

Funding
The authors received no financial support for the research and/or authorship of this article.

References

1) Alberg AJ, Ford JG, Samet JM. Epidemiology of lung cancer: ACCP evidence-based clinical practice guidelines (2nd edition). Chest 2007;132:29-55.

2) Scott WJ, Gobar LS, Terry JD, Dewan NA, Sunderland JJ. Mediastinal lymph node staging of non-small-cell lung cancer: a prospective comparison of computed tomography and positron emission tomography. J Thorac Cardiovasc Surg 1996;111:642-8.

3) Silvestri GA, Tanoue LT, Margolis ML, Barker J, Detterbeck F. The noninvasive staging of non-small cell lung cancer: the guidelines. Chest 2003;123:147-56.

4) Özgüven MA, Öztürk E. Genel prensipler ve uygulama alanları. Akciğer Kanseri PET El Kitabı; 2008. s. 1-77.

5) Reed CE, Harpole DH, Posther KE, Woolson SL, Downey RJ, Meyers BF, et al. Results of the American College of Surgeons Oncology Group Z0050 trial: the utility of positron emission tomography in staging potentially operable non-small cell lung cancer. J Thorac Cardiovasc Surg 2003;126:1943-51.

6) De Leyn P, Lardinois D, Van Schil PE, Rami-Porta R, Passlick B, Zielinski M, et al. ESTS guidelines for preoperative lymph node staging for non-small cell lung cancer. Eur J Cardiothorac Surg 2007;32:1-8.

7) Freixinet Gilart J, García PG, de Castro FR, Suárez PR, Rodríguez NS, de Ugarte AV. Extended cervical mediastinoscopy in the staging of bronchogenic carcinoma. Ann Thorac Surg 2000;70:1641-3.

8) Shin KM, Lee KS, Shim YM, Kim J, Kim BT, Kwon OJ, et al. FDG PET/CT and mediastinal nodal metastasis detection in stage T1 non-small cell lung cancer: prognostic implications. Korean J Radiol 2008;9:481-9.

9) Gould MK, Kuschner WG, Rydzak CE, Maclean CC, Demas AN, Shigemitsu H, et al. Test performance of positron emission tomography and computed tomography for mediastinal staging in patients with non-small-cell lung cancer: a meta-analysis. Ann Intern Med 2003;139:879-92.

10) Tasci E, Tezel C, Orki A, Akin O, Falay O, Kutlu CA. The role of integrated positron emission tomography and computed tomography in the assessment of nodal spread in cases with non-small cell lung cancer. Interact Cardiovasc Thorac Surg 2010;10:200-3.

11) Liu BJ, Dong JC, Xu CQ, Zuo CT, Le JJ, Guan YH, et al. Accuracy of 18F-FDG PET/CT for lymph node staging in non-small-cell lung cancers. Chin Med J (Engl) 2009;122:1749-54.

12) Andrea B, Pelossi E, Skanjeti A, Arena V, Errico E, Borasio P, et al Preoperative intrathoracic lymph node staging in patients with non-small-cell lung cancer: accuracy of integrated positron emission tomography and computed tomography. European Journal of Cardio-thoracic Surgery 2009;36:440-5.

13) Lee BE, Redwine J, Foster C, Abella E, Lown T, Lau D, et al. Mediastinoscopy might not be necessary in patients with non-small cell lung cancer with mediastinal lymph nodes having a maximum standardized uptake value of less than 5.3. J Thorac Cardiovasc Surg 2008;135:615-9.

14) Kwan A, Seltzer M, Czernin J, Chou MJ, Kao CH. Characterization of hilar lymph node by 18F-fluoro-2- deoxyglucose positron emission tomography in healthy subjects. Anticancer Res 2001;21:701-6.

15) Kang WJ, Chung JK, So Y, Jeong JM, Lee DS, Lee MC. Differentiation of mediastinal FDG uptake observed in patients with non-thoracic tumours. Eur J Nucl Med Mol Imaging 2004;31:202-7.

16) Cerfolio RJ, Bryant AS. Ratio of the maximum standardized uptake value on FDG-PET of the mediastinal (N2) lymph nodes to the primary tumor may be a universal predictor of nodal malignancy in patients with nonsmall-cell lung cancer. Ann Thorac Surg 2007;83:1826-9.

17) Kim BT, Kim Y, Lee KS, Yoon SB, Cheon EM, Kwon OJ, et al. Localized form of bronchioloalveolar carcinoma: FDG PET findings. AJR Am J Roentgenol 1998;170:935-9.

18) Al-Sarraf N, Aziz R, Gately K, Lucey J, Wilson L, McGovern E, et al. 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 2008;33:104-9.

19) Lee PC, Port JL, Korst RJ, Liss Y, Meherally DN, Altorki NK. Risk factors for occult mediastinal metastases in clinical stage I non-small cell lung cancer. Ann Thorac Surg 2007;84:177-81.