IMPACT OF DUAL TIME POINT FDG PET IMAGING IN THE ASSESSMENT OF SOLITARY PULMONARY NODULES

Objective: The aim of this study was to assess the diagnostic potential of ¹⁸F-FDG Positron Emission Tomography imaging in the evaluation of patients with solitary pulmonary nodules, by comparing the diagnostic criteria in single time point imaging method to two different diagnostic criteria in dual-time-point imaging.

Methods: This retrospective study was conducted in the Hospital of the University of Pennsylvania, and data was collected and analyzed in the period from September 2005 to March 2006, from the pooled hospital studies for the last eight years. Two hundred sixty five patients were included (161 men, 104 women, age range: 41–92 years). All had solitary pulmonary nodules on computed tomography, and the diagnosis was confirmed by biopsy or by follow up computed tomography. All 265 patients underwent whole body FDG PET scan, and 255 of them had PET scan two time points. The maximum standardized uptake values of nodules were calculated for both time points. On single time point imaging we set the maximum standardized uptake value of 2.5 as a cutoff criterion for malignancy. On dual time point imaging, first criterion of malignancy was set as any increase in the maximum SUV from the first to second time point. The second criterion was set as either no change or increase in the maximum standardized uptake value between the two time points. Sensitivity, specificity and accuracy were calculated for the three methods by using the biopsy results and clinical follow up as gold standard.

Results: Biopsy and follow-up revealed 72 patients with malignant lung nodules, whereas 193 patients had benign nodules. Single time point imaging with a threshold maximum standardized uptake value of 2.5 had a sensitivity, specificity and accuracy of 63%, 92% and 85% respectively. On dual-time-point imaging, for the initial criterion for malignancy, the sensitivity, specificity and accuracy were 81%, 95% and 91% respectively. On dual time point imaging, for the second criterion for malignancy, the sensitivity, specificity and accuracy were 92%, 93%, and 92% respectively.

Conclusion: Dual-time-point FDG PET imaging using both criteria has higher sensitivity, specificity and accuracy compared to single time imaging. Dual-time-point FDG PET imaging should be included in the clinical workup of patients with pulmonary nodule.

Key Words: Dual-time-point imaging, FDG-PET, Solitary pulmonary nodule

JRMS August 2008; 15(2): 6-14

Introduction

Solitary pulmonary nodules are defined as focal, round or oval areas of increased opacity in the lung that measure less than three cm in diameter.^(1,2,3) An estimated 150,000 solitary pulmonary nodules are detected annually in the United States and are often discovered incidentally at chest radiography or computed tomography (CT).^(3,4) These nodules are caused by a variety of disorders including neoplasms, infection, inflammation, and vascular and congenital abnormalities. Although most solitary pulmonary nodules have benign causes, 30%–40% of these nodules are malignant.^(4-6)

¹⁸F-FDG has established role in oncology, which includes initial diagnosis, staging, and therapeutic follow-up studies.^(7,8)  Despite its proven utility, the application of PET is limited by its variable sensitivity and specificity estimates. One of the main reasons for this limitation is that many inflammatory lesions also have elevated ¹⁸F-FDG uptake in PET, leading to false-positive results.^(9,10)  On the other hand, some types of cancers, for example, carcinoid tumor and bronchoalveolar carcinomas, have low ¹⁸F-FDG uptake below the diagnostic threshold for ¹⁸F-FDG uptake in malignant lesions.^(11,12)

A maximum standardized uptake value (SUV) of 2.5 as a cutoff criterion for malignancy has been used for diagnosing pulmonary malignancies with ¹⁸F-FDG PET.^(13-20)  However, one study indicated that the sensitivity of this SUV cutoff was lower than that of visual assessment.⁽²¹⁾ Some authors have recommended using visual evaluation rather than the SUV for small solitary pulmonary nodules,⁽²²⁾ suggesting that the classical SUV criterion of 2.5 is inappropriate for diagnosing malignancies with low ¹⁸F-FDG uptake.^(21,23) Studies have shown that the uptake of ¹⁸F-FDG continues to increase in malignant tumors for several hours after ¹⁸F-FDG injection.^(10,12) It has been deduced that this difference in the time course of ¹⁸F-FDG uptake could be used to improve the ability of PET to distinguish benign lesions from malignant lesions. Preliminary studies have been performed using FDG PET with dual-time-point imaging on head and neck cancers, breast cancer and malignant lung lesions. Those results demonstrated significant improvement in the diagnostic accuracy of FDG PET scan.^(12,21-25)

On the basis of the promising results from dual-time-point imaging research, the present study was undertaken to assess whether dual-time-point acquisition can improve the diagnostic utility of PET in Solitary Pulmonary nodules.

Methods

This retrospective study was conducted in the Hospital of the University of Pennsylvania, and data was collected and analyzed in the period from September 2005 to March 2006, from the pooled hospital studies for the last five years. Two hundred sixty five patients (161 men, 104 women; mean age 67 years; age range: 41–92 years) were included in this retrospective analysis. All patients had suspected solitary pulmonary nodules detected by CT. All our study patients had ¹⁸F-FDG PET and CT scanning acquired in two different occasions, with a time gap (0-35 days).  All patients underwent whole body PET scans, and 255 patients were examined twice: initial whole-body imaging followed by a second scan for the chest only. Informed consent  was obtained from all patients. At the time of ¹⁸F-FDG injection all patients had fasted for at least four hours and had blood sugar levels of <150 mg/dL.

Image acquisition for the whole-body scan started at a mean time point of 60 minutes after injection of 2.52 MBq/kg of body weight. This first scan (scan A) included neck, thorax, abdomen, pelvis and upper thighs. It consisted of four or five emission frames of 25.6-cm length with an overlap of 12.8 cm covering an axial length of 64–76.8 cm, including six to seven beds and duration of the scan was 18-21 minutes. A second emission scan of the thorax only (scan B) was acquired on 255 patients at a mean time of 110 minutes after tracer injection (range 100–120 min), including two beds and duration of scan was ranging between six minutes. A transmission scan was obtained with both sets of images for attenuation correction.  Image reconstruction was performed with an iterative ordered-subsets expectation maximization algorithm with four iterations and eight subsets. Attenuation-corrected images were obtained by applying transmission maps, which were acquired after ¹⁸F-FDG injection with a ¹³⁷Cs source interleaved with the emissions scans.

Regions of interest (ROIs) were overlaid onto the lesions on fully corrected PET images of scans A and B axial slices (Fig. 1). This was achieved by direct visual assessment of the lesion position on the CT scan and subsequent identification of the corresponding area on PET scans A and B.

  SUV=1.8                                                                                SUV=3.2

Fig.1. An axial slices region of interest (ROI) were placed around the lesion on first time (A) and second time (B) images, in order to calculate the maximum SUV

Fig. 2. Dual-time-point FDG PET imaging of  60 years old  man with 1.8 cm solitary pulmonary nodule in the left lung on transverse, sagittal and axial slices. First time whole body image (A) shows a focal area with FDG upatke (arrow) and with SUV= 2.5. Second time image of the chest(B) shows more prominant FDG upatke with SUV=2.9. Pathological diagnosis of this nodule was moderately differentiated adenocarcinoma

Results

Of the 265 nodules included in this study, 72 (27%) proved to be malignant and 193 (73%) benign. The sizes of these 265 nodules were as follows: 32 nodules <1 cm in diameter, 57 nodules 1-1.5 cm and 176 nodules 1.6-3 cm. Diagnoses of 158 nodules were decided to be benign according to the clinical follow up data.

On single time point imaging, 60 nodules had SUV≥2.5 (45 malignant and 15 benign), 205 nodules had SUV<2.5 (27 malignant and 178 benign). While 119/205 nodules did not show any visually apparent FDG uptake in the first scan and SUVs of those lesions were equal to the lung background.

On dual-time-point imaging, 60 nodules had increased in SUV between scan A and scan B (50 malignant and 10 benign), 70 nodules had drop in SUV (three malignant and 67 benign) and 11 nodules had no change in SUV (seven malignant and four benign). One hundred and fourteen nodules had no FDG uptake on scan A and B and the calculated SUV= lung background (two malignant and 112 benign). Figures 2 & 3 show dual-time-point imaging in two patients with malignant pulmonary nodules, including SUV change and histopathology diagnosis.

When assessing the diagnostic value of the first emission  scan  by applying an SUV threshold of 2.5

for separating benign from malignant lesions, sensitivity, specificity and accuracy were 63%, 92% and 84% respectively. When applying this criterion to small nodules of less than 1cm, the values were 50%, 90% and 75% respectively compared to 65%, 93% and 85% for larger nodules.

When any increase in SUV between the first and second scans was applied as a criterion for malignancy, the sensitivity, specificity and accuracy were 83%, 95% and 91% respectively. When any increase or no change in SUV between the first and second scans criterion was applied, the values were 92%, 93%, and 92% respectively. Table I and II shows SUV1, SUV2 and histopathological diagnoses for benign and malignant lung nodules. Table III shows the sensitivity, specificity, accuracy, NPV and PPV in the three criteria used in our study according to single time SUV and dual time change.

Discussion

In addition to visual assessment of the metabolic activity of the nodules, measurement of the SUV for the semiquantitative assessment of ¹⁸F-FDG uptake in pulmonary lesions has proven to assist in differentiating between malignant and benign nodules.^(13-20) Several reports consider it to be a simple and useful tool for this purpose, and most publications conclude that a threshold value of 2.5 is optimal for obtaining a high sensitivity while maintaining a good specificity.^(13-20) However, several reports and observations on the day-to-day clinical practice indicate that a significant degree of overlap exists between the uptake values of benign and malignant lesions.^(16-18) Certain inflammatory lesions, including granulomatous processes, fungal infections, or bacterial infections, can be noted with SUVs of >2.5,^(15,26) thereby limiting specificity of this method. Most inflammatory lesions would fall below the 2.5 SUV thresholds, whereas the majority of malignant lesions would have high SUVs. However,  small  malignant  lesions may have under

estimated SUV due to partial volume effect and limited resolution of PET scanner.^(27-28) On the other hand, carcinoid tumor and bronchoalveolar carcinomas can have low levels of ¹⁸F-FDG uptake, and the SUV in such tumors may fall below the 2.5 limit for malignancy in this criterion.^(11,12) Those factors have potential effect on the diagnostic accuracy of using this criterion, because it can lead to misinterpretation of malignant lesions into benign ones.

In our study, when adopting the SUV ≥ 2.5 as criterion for malignancy, the sensitivity, specificity and accuracy were 63%, 92% and 84% respectively. Our results have lower sensitivity and comparable specificity compared to those published in the literature.^(13-18) When applying this criterion to small nodules of less than 1cm, the values were 50%, 90% and 75% compared to 65%, 93% and 85% for larger nodules. This can show the limitation of this criterion induced primarily by underestimation of SUV due to impact of partial volume effect.

Dual-time-point FDG PET imaging was suggested as discriminator of benign and malignant diseases, with images being obtained at one and two hours after the administration of ¹⁸F-FDG.  Hustinx et al.⁽⁹⁾ had acquired dual time point imaging for head and neck tumors, and he used a threshold of 10% increase in measured values. He reported higher sensitivity (100% vs. 80%), while maintaining an excellent specificity (89% vs. 94%), than that obtained from a single image acquisition using the usual SUV threshold method.   Zhuang et al.⁽¹⁰⁾ found that malignant lesions showed a significant increase in SUV over time and that benign lesions showed a decrease over time. Lodge et al.⁽²⁶⁾ came to a similar conclusion in a study of 29 patients with various benign and malignant soft-tissue masses. Rakesh et al.⁽²⁵⁾ had applied dual time point imaging in breast cancer, and found that breast malignancies show increasing FDG uptake with time, whereas the uptake of ¹⁸F-FDG in inflammatory lesions and normal breast tissues decreases over time. 

Table I. SUV1 and SUV2 in malignant lung nodules

Table II. SUV1 and SUV2 in benign lung nodules

Table III. The results of statistical analysis in three different methods used in the assessment of solitary pulmonary nodules