Eurasian Journal of Pulmonology

ORIGINAL ARTICLE
Year
: 2020  |  Volume : 22  |  Issue : 1  |  Page : 23--28

Retrospective analysis of false positive ratio of our patients with lung cancer at positron emission tomography-CT screen


Hasan Oguz Kapicibasi1, Pınar Mutlu2, şahınur Aycan Alkan2, Nihal Arzu Mirici2, Buse Yuksel3, Çoşkun Bakar3,  
1 Department of Thoracic Surgery, Çanakkale 18 Mart University, Çanakkale, Turkey
2 Department of Chest, Çanakkale 18 Mart University, Çanakkale, Turkey
3 Department of Public Health, Çanakkale 18 Mart University, Çanakkale, Turkey

Correspondence Address:
Dr. Pınar Mutlu
Department of Chest, Çanakkale 18 Mart University, Çanakkale
Turkey

Abstract

BACKGROUND: In lung cancer, staging is necessary to give the best treatment to the patient and to estimate the best prognosis. The aim of this study was to compare the pathology results of the lung masses and mediastinal lymph nodes and to evaluate the sensitivity and specificity values of positron emission tomography.computerized tomography (PET.CT) and to determine the maximal threshold of maximum standardized uptake volume (SUVmax). MATERIALS AND METHODS: We retrospectively evaluated the PET.CT SUVmax values and pathology results of the patients who had a mass, mediastinal lymph node, or scalene lymph node in our patients between 2016 and 2018. RESULTS: Fifty.one people and 75 pathology materials were included in our study. We used the receiver operating characteristic curve analysis to determine the cutoff value for SUVmax value and calculated the cutoff value as 6.65. In our study, the sensitivity and specificity were calculated as 63% and 71%, respectively. We calculated the positive predictive value as 73.5% and the negative predictive value as 61%. CONCLUSION: As a result, considering the common inflammatory and granulomatous diseases seen in our country, we concluded that benign diseases should be considered before malignancy in SUVmax value below 6.6. We continue to add new patients and new data to our study to find the most appropriate threshold of SUVmax value for the health values of our country.



How to cite this article:
Kapicibasi HO, Mutlu P, Alkan &A, Mirici NA, Yuksel B, Bakar &. Retrospective analysis of false positive ratio of our patients with lung cancer at positron emission tomography-CT screen.Eurasian J Pulmonol 2020;22:23-28


How to cite this URL:
Kapicibasi HO, Mutlu P, Alkan &A, Mirici NA, Yuksel B, Bakar &. Retrospective analysis of false positive ratio of our patients with lung cancer at positron emission tomography-CT screen. Eurasian J Pulmonol [serial online] 2020 [cited 2020 Oct 27 ];22:23-28
Available from: https://www.eurasianjpulmonol.com/text.asp?2020/22/1/23/283635


Full Text



 Introduction



Lung cancer is the most common type of cancer in the world for many years. In 2012, it is estimated that there are 1.8 million new cases worldwide. The disease is the most common type of cancer in men worldwide (1.2 million, 16.7% of the total). Its incidence in Central and Eastern Europe is 53.5/100,000. Although the rates are slightly lower in women, it is mostly seen in North America (%0,0033.8) and in northern European countries(%0,0023.7).

As well as its frequency, lung cancer increases its severity with its mortality. Lung cancer is the most common cause of death of cancer worldwide and is estimated to be responsible for one in five (1.59 million deaths, 19.4% of the total).[1]

Because of respiratory tract cancer in Turkey it is 75,993 patients have been lost their lives in 2014. This constitutes 31.1% of all cancer-related deaths.[2]

Even though the continuous development of lung cancer treatment, unfortunately, its prognosis is still very poor.

According to the studies, prognosis is associated with the stage of clinical diagnosis; the 5-year survival rate is 38%–67% in Stage 1, but only 1% in Stage 4.[3] If patients with lung cancer accept surgery at an early stage, the 10-year survival rate can be 88%.[4] For this reason, early diagnosis and timely treatment is very important in patients with lung cancer.

Diagnosis of the disease is often based on pathological examination with appropriate method (bronchoscopic biopsy/transthoracic fine-needle aspiration biopsy/surgery) after the detection of mass by computed tomography.

For patients, correct staging is essential to obtain the most effective treatment and to estimate the best prognosis. Using floro-2-deoxy-glucose (FDG), both computed tomography (CT) and positron emission tomography (PET) play an important role in the diagnosis and staging of lung cancer. In addition to mediastinal lymph node metastases, FDG-PET is highly sensitive in detecting extrathoracic metastases.[5]

PET, based on the fact that malignant cells have a higher rate of glycolysis than most of the surrounding normal cells. Glucose is also rapidly metabolized by tissues involved in granulomatous or inflammatory processes, and therefore, there will be some false positive results.[6] One of the reasons for false positivity is tuberculosis (TB), which is a common disease in the world and in our country.

Frequency of TB was reported as 17.3/100,000 in Turkey.[7]

Therefore, it is too much important for patient and health management to be careful, in terms of false positives that may increase during PET-CT use in staging lung cancer. In our study, we aimed to evaluate the sensitivity and specificity of PET-CT and to calculate the best threshold of SUVmax by comparing the pathology results of the mediastinal lymph nodes.

 Materials and Methods



Working group

We retrospectively evaluated the PET-CT FDG results of the patients who had a mass in the lung, mediastinal lymph node, or scalene lymph node between 2016 and 2018. We compared the pathological manifestations with the samples that were available for tissue diagnosis (mediastinoscopy/lobectomy/segmentectomy/pneumonectomy/scalar lymph node biopsy).

Our study complies with the Declaration of Helsinki.

Fluorodeoxyglucose positron emission tomography-CT examination

All body scannings were performed by Siemens Biograph DUO PET/BT (USA). Blood glucose levels were adjusted to 150 mg/dl 6 h before screening. The scannings were performed using intravenous contrast medium for covering the vertex and upper thigh level while the patient was in supine position. The Farmasotik dose was taken to be 7–8 mCi position number 8 position duration 3 min. For attenuation correction and anatomic accuracy, a low-dose cross-section thickness of 3 mm CT was evaluated together.

Image examination and analysis

Image analysis was performed by a nuclear medicine specialist. The standardized uptake values (SUVs) were obtained by the calculation of the amount of FDG by body weight and by the corrected calculation of regional attenuation in the target tissue. Mediastinal lymph nodes were considered positive when the involvement was higher than intravenous involvement and was named according to lymph nodes Mountain and Dresler lymph node map.

Histopathological lymph node sample and analysis

Sampling was performed by the same surgeon who had knowledge about PET-CT results through lobectomy, segmentectomy, pneumonectomy, mediastinoscopy, or lymph node dissection. Mediastinal lymph nodes were performed according to the Mountain and Dresler lymph node map according to the naming of PET-CT. The chronic inflammatory process is called granulomatous changes and anthracosis benign pathology; adenocarcinoma, squamous cell carcinoma, bronchoalveolar cancer, small-cell lung cancer, and metastases of extrapulmonary malignancies were collected in the malignant group.

Data analysis

The results were obtained by comparing the results of PET-CT and histopathological examination. According to the results of PET-CT, FDG involvement was detected as false positive in the absence of malignancy in histopathological examination; lesions with FDG involvement and histopathologically diagnosed as malignant were positive; false negative results with no FDG involvement but with histopathological features; histopathologically, nonmalignant and noninvolvement PET-CT studies have been evaluated as true negative.

Statistical analysis

Data were analyzed with SPSS Package Program version 19.0 (IBM Co., Somers, NY, USA). In the presentation of descriptive data, number, percentage, mean, standard deviation, median, minimum, and maximum were used. Receiver operating characteristic (ROC) curve analysis was used to determine the threshold value of SUVmax for use in the differentiation of benign malignant. According to the result of analysis of benign malignancy, SUVmax cutoff value was accepted as 6650. Sensitivity, specificity, positive predictive value, negative predictive value, and accuracy were calculated according to this value. P < 0.05 was accepted for statistical significance.

 Results



Nine (17.6%) female and 42 (82.4%) male patients were included in the study. The researches included were 75 pathological materials, of which 32.0% (n = 24) were mass and 68.0% (n = 51) were lymph nodes. 15.7% (n = 8) of the lymph nodes, right upper paratracheal (2R), 35.3% (n = 18) right lower paratracheal (4R), 23.5% (n = 12) subcarinal,[7] 5.9% (n = 3) left upper paratracheal (2 L), 9.8% (n = 5) left lower paratracheal (4 L) and 9.8% (n = 5) was located in the supraclavicular region [Table 1].{Table 1}

The average of SUVmax was 7.4 ± 4.7, and the median was 5.6 (min: 0.0 and max: 25.5).

ROC curve analysis was used to determine the threshold value of SUVmax in the distinction of benign–malignant pathology. The analyses included were 41 malignant[1] and 34 benign[2] pathology [Table 2], [Table 3], [Table 4]. The area under the curve in the ROC curve analysis was 0.637 (95% confidence interval [CI]: 0.504–0.771) (P = 0.042). According to the result of the analysis of benign and malignant, SUVmax estimation value was accepted as 6650 [Figure 1].{Table 2}{Table 3}{Table 4}{Figure 1}

 Discussion



Lung cancer is the most common cancer in men in Turkey and its incidence rate is 52.5/100,000. This rate is 8.7/100,000 in women. In lung cancer, more than half of the patients are diagnosed at advanced stage.[8]

Early diagnosis is vital in lung cancer. Although CT provides three-dimensional imaging of the lungs, it does not provide information about physiological/metabolic features. PET goes beyond anatomical imaging to enable the characterization and measurement of biological processes at the cellular level. Combined PET-CT technology provides the clinician with clear information about where healthy lung tissue terminates and where tumor tissue begins.[9]

PET-CT is mainly based on the fact that malignant cells will consume more glucose than normal tissue. The most commonly used radiopharmaceutical agent is the glucose analog used to monitor glucose transport and metabolism 2-(fluorine-18 [18F])-floro-2-deoksi-d-glucose (FDG), where 18F is a positron spreader that creates the high-energy photons. The rate of cellular glycolysis is reflected by the degree of FDG involvement and can be determined from the correction data by imaging data, so that the photons are not attenuated by body tissues.[10]

PET-CT is also of great benefit for staging, especially in nonsmall-cell lung cancer (NSCLC). The sensitivity and specificity of the CT evaluation performed to identify lymph node metastasis in a rectopposed study by Silvestri et al.[11] were 51% (95% CI: 47%–54%) and 85% (95% CI: 84%–88%), respectively. The same values for PET-CT were 74% (95% CI: 69%–79%) and 85% (95% CI: 82%–88%), respectively. Again in the same study, false positive rate was 15%–20% and false negative rate was 10%–15%. Most of the false positives are inflammatory diseases. Because of the inactive TB lymph nodes that are not bred by live TB, it is difficult to distinguish between noncalcified lymph node TB and metastatic lymph nodes.[12] Kang et al.[13] suggested that 18F-FDG showed low diagnostic sensitivity in the differentiation of NSCLC and lung TB. However, Shaw et al.[14] have argued that PET-CT is a valuable method to exclude mediastinal lymph node involvement in NSCLC, even in a high TB-endemic region, and that the PET-CT positive results should not necessarily exclude potential TB stated that they should. In addition, granulomatous diseases such as sarcoidosis may mimic malignant diseases with mediastinal lymph node involvement in PET-CT.[15] Anthrax in the lymph nodes may be associated with malignancy and TB, or it may cause high FDG PET involvement alone.[16]

The overlap between the standardized uptake value (SUV) in malignant and benign lesions has led to the investigation of several dichotomization methods, such as the use of SUV cutoff thresholds, dual tracer imaging, dual time point imaging, or delayed imaging. There is, however, no consensus about the use of18 F-FDG PET to differentiate TB from malignancy or other granulomatous or other inflammatory lesions.[17]

In another study with 87 histologically confirmed patients with lung malignancy and 46 patients with histologically confirmed TB lesions, 1st h SUV and 2nd h SUV values of malignant lung lesions were significantly higher than TB lesions.[18]

In countries such as India, Indonesia, and China, where TB is endemic, improving control of TB can help improve the diagnostic accuracy of PET-CT in lung cancer. Despite the decrease in frequency in recent years in Turkey, in terms of TB continues to take place in the list of priority countries of the World Health Organization. 13,378 TB cases of TB in Turkey entered the record in 2014. The rate of the cases was 19.8 in 100,000 in men and 14.6 in 100,000 in women.[19]

In Western countries, the prevalence of SUVmax of 2.5 is considered to be a cutoff value for benign and malignant lesions.[20] However, using the same value in countries where TB is endemic reduces the diagnostic value of PET-CT in lung cancer.

According to a study conducted by Goo et al.[21],[22] in South Korea, increased SUVmax values in the focal pulmonary lesions were observed above the 2.5% of the TB-max values and in the study performed by Kumar et al. The sensitivity and specificity values of PET-CT were 87% and 70%, respectively. Shaw et al.[14] reported that an SUVmax cutoff of 4.5 could increase diagnostic accuracy from 64.0% to 84.7% compared to a cutoff of 2.5. In a retrospective study of 75 pathological specimens, we found that the mean SUVmax values of PET-CT were 7.4 ± 4. We used the ROC curve analysis to determine the cutoff value for SUVmax value in benign–malignant, and we calculated the cutoff value as 6.65. In our study, the sensitivity and specificity were 63% and 71%, respectively. We calculated the positive predictive value as 73.5% and the negative predictive value as 61%.

In countries with widespread TB, progressing to TB control and improving the threshold value of PET-CT may be promising to reduce the false positive PET-CT ratio in lung cancer.

The limitations of our study were retrospective file scanning and the lack of sample number.

 Conclusion



Although our study was performed on the basis of the retrospective data of 51 patients and 75 pathological materials, we concluded that benign diseases should be considered before malignancy in the SUVmax value below 6.6 when considering the common inflammatory and granulomatous diseases commonly seen in our country. We continue to add new patients and new data to our study to find the most appropriate threshold value for our country's health values.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

References

1Globocan 2012: estimated Cancer Incidence Mortality and Prevalence Worldwide in 2012-who. Available from: http://globocan.iarc.fr/Pages/fact_sheets_cancer.aspx?cancer=lung. [Last accessed on 2020 Jan 08].
2TUIK Death Reason Statistics; 2014. Available from: http://www.tuik.gov.tr/PreHaberBultenleri.do?id=18855. [Last accessed on 2020 Jan 08].
3Jemal A, Siegel R, Ward E, Murray T, Xu J, Thun MJ, et al. Cancer statistics, 2007. CA Cancer J Clin 2007;57:43-66.
4Mazzone P, Obuchowski N, Mekhail T, Meziane M, Ahmad M. Lung cancer screening: Is it time for a change in policy? Cleve Clin J Med 2007;74:441-8.
5Steinert HC. PET and PET-CT of lung cancer. In: Juweid M. Hoekstra O, editors. Positron Emission Tomography. Methods in Molecular Biology (Methods and Protocols). Vol. 727. Humana Press; 2011. Available from: https://link.springer.com/protocol/10.1007%2F978-1-61779-062-1_3. [Last accessed on 2020 Jan 08].
6Roberts PF, Follette DM, von Haag D, Park JA, Valk PE, Pounds TR, et al. Factors associated with false-positive staging of lung cancer by positron emission tomography. Ann Thorac Surg 2000;70:1154-9.
7Tuberculosis Surveillance and Monitoring in Europe; 2016. Available from: http://www.euro.who.int/__data/assets/pdf_file/0005/310100/TB-surveillance-report-2016-Turkey.pdf. [Last accessed on 2020 Jan 08].
8Public Health Agencies Cancer Statistics for Turkey; 2017. Available from: https://hsgm.saglik.gov.tr/depo/birimler/kanser-db/istatistik/2014-RAPOR._uzuuun.pdf. [Last accessed on 2020 Jan 08].
9Cancer Treatment Centers of America. Available from: https://www.cancercenter.com/lung-cancer/pet-ct-scan/. [Last accessed on 2020 Jan 08].
10Al-Jahdali H, Khan AN, Loutfi S, Al-Harbi AS. Guidelines for the role of FDG-PET/CT in lung cancer management. J Infect Public Health 2012;5 Suppl 1:S35-40.
11Silvestri GA, Gould MK, Margolis ML, Tanoue LT, McCrory D, Toloza E, et al. Noninvasive staging of non-small cell lung cancer: ACCP evidenced-based clinical practice guidelines (2nd edition). Chest 2007;132:178S-201S.
12Sathekge MM, Maes A, Pottel H, Stoltz A, van de Wiele C. Dual time-point FDG PET-CT for differentiating benign from malignant solitary pulmonary nodules in a TB endemic area. S Afr Med J 2010;100:598-601.
13Kang F, Wang S, Tian F, Zhao M, Zhang M, Wang Z, et al. Comparing the diagnostic potential of 68Ga-alfatide II and 18F-FDG in differentiating between non-small cell lung cancer and tuberculosis. J Nucl Med 2016;57:672-7.
14Shaw JA, Irusen EM, von Groote-Bidlingmaier F, Warwick JM, Jeremic B, du Toit R, et al. Integrated positron emission tomography/computed tomography for evaluation of mediastinal lymph node staging of non-small-cell lung cancer in a tuberculosis-endemic area: A 5-year prospective observational study. S Afr Med J 2015;105:145-50.
15Kumar A, Dutta R, Kannan U, Kumar R, Khilnani GC, Gupta SD, et al. Evaluation of mediastinal lymph nodes using F-FDG PET-CT scan and its histopathologic correlation. Ann Thorac Med 2011;6:11-6.
16Prabhakar HB, Rabinowitz CB, Gibbons FK, O'Donnell WJ, Shepard JO, Aquino SL. Imaging features of sarcoidosis on MDCT, FDG PET, and PET/CT. Am J Roentgenol 2008;190:1-6. Available from: https://www.ajronline.org/doi/pdf/10.2214/AJR.07.7001. [Last accessed on 2020 Jan 08].
17Ankrah AO, van der Werf TS, de Vries EF, Dierckx RA, Sathekge MM, Glaudemans AW. PET/CT imaging of Mycobacterium tuberculosis infection. Clin Transl Imaging 2016;4:131-44.
18Cho J, Kim S, Choe JG, Eo JS, Rhee S, Choi S. Delayed F-18 FDG uptake in PET distinguishes between TB and lung cancer: Determination of the optimal cut-off level using ROC analysis. J Nucl Med 2016;57 Suppl 2:1469.
19Available from: https://hsgm.saglik.gov.tr/depo/haberler/verem-savas-raporu-2016-2017/Turkiyede_Verem_Savasi_2016_Raporu.pdf. [Last accessed on 2020 Jan 08].
20Knight SB, Delbeke D, Stewart JR, Sandler MP. Evaluation of pulmonary lesions with FDG-PET. Comparison of findings in patients with and without a history of prior malignancy. Chest 1996;109:982-8.
21Goo JM, Im JG, Do KH, Yeo JS, Seo JB, Kim HY, et al. Pulmonary tuberculoma evaluated by means of FDG PET: Findings in 10 cases. Radiology 2000;216:117-21.
22Elri T, Aras M, Salihoglu YS, Erdemir RU, Cabuk M. A potential pitfall in the use of 68Ga-PSMA PET/CT: Anthracosis. Rev Esp Med Nucl Imagen Mol 2017;36:65-6.