Evidence-Based PET for Thoracic Tumours

Fluorine-18 fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) is a robust imaging tool that is currently used in daily clinical practice for the evaluation of thoracic malignancies. This chapter provides an overview of the current evidence-based data on the usefulness of PET/CT for the evaluation of patients with thoracic tumours including lung cancer, pleural and thymic tumours, and esophageal cancer.


Introduction
Fluorine-18 fluorodeoxyglucose positron emission tomography/computed tomography ( 18 F-FDG PET/CT) is a robust imaging tool that is currently used in daily clinical practice for the evaluation of thoracic malignancies. This chapter provides an overview of the current evidencebased data on the usefulness of PET/CT for the evaluation of patients with thoracic tumours including lung cancer, pleural and thymic tumours, and esophageal cancer.

Evidence-Based Data on PET in Primary Lung Tumours
Herein we reviewed recent evidence-based data on the usefulness of 18 F-FDG PET/CT for: (1) characterization of solitary pulmonary nodules (SPNs), (2) non-small cell lung cancer (NSCLC) staging, (3) restaging after induction therapy and systemic therapy response assessment in NSCLC, (4) radiation therapy planning, (5) diagnosis of lung cancer recurrence in NSCLC, (6) prognostic evaluation, (7) management of small cell lung cancer (SCLC).

Characterization of Solitary Pulmonary Nodules (SPNs)
Characterizing a SPN detected incidentally or, as is the case more recently, on CT screening for lung cancer is a major public health issue. 18 F-FDG PET/CT is not indicated for characterization of SPNs of less than 8 mm in diameter according to current guidelines [1]. This threshold was set to take into account the spatial resolution of PET systems, due to the significant risk of false-negative findings for small lesions. However, over the last decade, the spatial resolution of 18 F-FDG PET/CT has significantly increased and future analysis could verify if this threshold will be modified accordingly.

Single-Time-Point 18 F-FDG PET or PET/CT
In the last decade, a robust evidence has been produced on the potential use of 18 F-FDG PET/ CT in early diagnosis of lung cancer (Table 5.1). Chien and colleagues [2] in 2013 conducted a systematic review on this topic reporting evidence of lung cancer screening programmes with 18 F-FDG PET, in which the estimated pooled sensitivity and specificity were 83% and 91%, respectively. At that moment, despite PET appeared to have high sensitivity and specificity as a selective screening modality, the role of primary PET screening for lung cancer remained unknown and still undefined. Subsequently, a further systemic analysis [3] reported a very high (98.7%) pooled sensitivity of 18 F-FDG PET/CT in this setting while specificity was suboptimal (58.2%).
In 2016, the research team headed by Madsen [4] suggested that 18 F-FDG PET/CT can rule out malignancy in most SPNs due to high sensitivity (recommendation level A) but at the same time the sensitivity of 18 F-FDG PET/CT in general is insufficient to rule out mediastinal lymph node metastasis (recommendation level A). Therefore, with few exceptions (lesions <1 cm and non-solid lesions), they concluded that SPNs could be presumptively considered benign if 18 F-FDG PET is negative. In addition, lymph node metastasis in the mediastinum cannot be ruled out on the basis of a negative 18 F-FDG PET/CT, and confirmative (mini)invasive staging should be performed in most patients.
More recently, a further meta-analysis [5] showed that the pooled sensitivity and specificity of 18 F-FDG PET/CT in characterizing SPNs were 82% and 81%, respectively, demonstrating moderate accuracy for 18 F-FDG PET/CT in differentiating malignant from benign SPNs.
A further meta-analysis exploring the value of 18 F-FDG PET/CT in the diagnosis of SPNs was reported in 2018 [6]. Pooled results indicated a sensitivity of 89% and a specificity of 70%. Considering the unsatisfactory results, especially in terms of specificity, the authors stated that 18 F-FDG PET/CT cannot replace the "gold standard" pathology by resection or biopsy. Not dissimilar results have been reported in a further recent meta-analysis performed by Divisi and co-workers [7]. The authors concluded that despite 18 F-FDG PET/CT presents a fairly good diagnostic accuracy in SPNs evaluation, it should not be considered as a discriminatory test rather than a method to be included in a clinical and diagnostic pathway.
Interestingly, Deppen and co-workers [8] evaluated the accuracy of 18 F-FDG PET in diagnosing lung cancer comparing populations with or without a risk for endemic infectious lung disease. They observed a 16% lower average adjusted specificity in regions with endemic infectious lung disease (61%) compared with non-endemic regions (77%). On the other hand, the sensitivity did not change appreciably by endemic infection status, even after adjusting for relevant factors. On the light of these results, the authors did not suggest the use of 18 F-FDG PET to diagnose lung cancer in regions with endemic pulmonary infections unless an institution achieves test performance accuracy similar to that found in non-endemic regions.
Lastly, a meta-analysis investigates the diagnostic performance of 18 F-FDG PET/CT compared with diffusion-weighted magnetic resonance imaging (DW-MRI) for distinguishing malignant versus benign SPNs [9]. DW-MRI had a pooled sensitivity and specificity of 83% and 91%, respectively, compared with 78% and 81%, respectively, for PET/CT. The authors concluded that the diagnostic performance of DW-MRI is comparable or superior to that of 18 F-FDG PET/ CT in the differentiation of malignant and benign pulmonary lesions.

Dual-Time-Point (DTP) PET
Several authors have also explored the potential use of a DTP 18 F-FDG PET in differentiating malignant from benign SPNs (Table 5.1). In 2012, a meta-analysis was performed by Lin and co-workers [10] exploring the diagnostic performance of both single-time-point (STP) and DTP 18 F-FDG PET techniques. Sensitivity was higher with DTP imaging at moderate levels of specificity. This potential advantage of DTP over initial STP scanning was diminished at higher levels of specificity. Although there was no clear evidence to support the routine use of DTP imaging with 18 F-FDG PET in the differential diagnosis of pulmonary nodules, the authors suggested as such technique may provide additional information in selected cases with equivocal results from initial scanning. Other meta-analyses [11][12][13] reported similar diagnostic accuracy among DTP and STP 18 F-FDG PET or PET/CT in the diagnosis of SPNs. According to these results, the additional value of DTP compared to STP 18 F-FDG-PET/ CT resulted to be questionable.

18 F-FLT PET for Evaluation of Pulmonary Lesions
The potential use of fluorine-18 fluorothymidine ( 18 F-FLT) PET in patients with pulmonary lesions was evaluated by two meta-analyses [14,15] ( Table 5.1), which showed that 18 F-FLT PET had a higher specificity but lower sensitivity compared to 18 F-FDG PET in the evaluation of SPNs. Therefore, the authors assumed that 18 F-FLT and 18 F-FDG together could add diagnostic confidence for pulmonary lesions.

NSCLC Staging
Nodal (N) and distant metastases (M) staging is one of the major prognostic factors of survival in NSCLC patients. Accurate staging of distant metastases is crucial, as the treatment strategy is directly dependent on tumour stage. Although many studies have been reported in the last decades evaluating the performance of 18 F-FDG PET/CT in lung cancer staging, the results among studies are still almost controversial.
Two meta-analyses were specifically limited to early-stage NSCLC cases. In detail, Wang and co-workers [19] found that the negative predictive value (NPV) of 18 F-FDG PET/CT for lymph nodal mediastinal metastases was 94% for T1 disease and 89% for T2 disease. Including both T1 disease and T2 disease, the NPV was 93% for mediastinal metastases and 87% for overall nodal metastases. Interestingly, adenocarcinoma histology type and high 18 F-FDG uptake in the primary lesion were associated with greater risk of occult nodal metastases.
Similarly, a second meta-analysis [20] focused on patients with resectable NSCLC revealed that 18 F-FDG PET/CT had a pooled sensitivity and specificity for N staging of 81.3% (95%CI: 70.2-88.9%) and 79.4% (95%CI: 70-86.5%), respectively. The authors assumed that accuracy of 18 F-FDG PET/CT in N staging was insufficient to allow management and strategy of care based on 18 F-FDG PET/CT findings alone.
Lastly, two meta-analyses compared 18 F-FDG PET/CT and DW-MRI for detection of mediasti-nal nodal metastases in NSCLC [22,23] reporting similar results in terms of diagnostic accuracy among these two imaging methods.
Concerning the latter issue, a comparative meta-analysis MRI and 18 F-FDG PET/CT for the diagnosis of brain metastases in NSCLC [27] revealed that MRI had higher sensitivity (77%) than 18 F-FDG PET/CT (21%) for the diagnosis of brain metastases.
Chang et al. [28] found a higher sensitivity and specificity of 18 F-FDG PET/CT compared to bone scintigraphy (BS) in detecting bone metastases from NSCLC. A further more robust metaanalysis [29] showed that 18 F-FDG PET/CT is a better imaging method in terms of sensitivity and specificity compared to MRI and BS for detecting bone metastases from NSCLC, with a pooled sensitivity and specificity of 92% (95%CI: 88-95%) and 98% (95%CI: 97-98), respectively.
Finally, the diagnostic performance of 18 F-FDG PET/CT in detecting adrenal metastases from NSCLC was recently evaluated by Wu and co-workers [30]. The pooled sensitivity and specificity of 18

Restaging After Induction Therapy and Prediction of Treatment Response
The ability to identify potential responders to induction treatment may improve patient selection or surgery and may help in the development of response criteria suitable for routine monitoring of response. By providing information on the metabolic activity of tumour cells, 18 F-FDG PET/ CT has become a powerful tool in assessing treatment response. Zhang and colleagues [31] performed a meta-analysis to evaluate the value of 18 F-FDG PET in predicting the pathological tumour response of lung cancer to induction therapy. The authors found that 18 F-FDG PET could play an important role in predicting nonresponders to induction therapy in cases of lung cancer: indeed, the pooled sensitivity, specificity, positive predictive value, and negative predictive value for PET-predicted response were 83% (95%CI: 76-89%), 84% (95%CI: 79-88%), 74% (95%CI: 67-81%), and 91% (95%CI: 87-94%), respectively. A recent evidence-based article assessed the use of 18 F-FDG PET/CT for both assessing the efficacy of treatment response and performing post-treatment follow-up of lung cancer [32]. PET metabolic response (PERCIST criteria) has been shown to be a better predictor of histopathologic response than anatomic response metrics (WHO and RECIST criteria). 18 F-FDG PET/CT was indicated for treatment response assessment when it is performed within 6 months from treatment completion, though evidence for its comparative effectiveness with chest CT is still evolving.

Radiation Therapy Pretreatment Planning in NSCLC
18 F-FDG PET/CT may also increase the likelihood of correctly delineating tumour tissue before radiotherapy dose planning. In 2017, Hallqvist and colleagues [33] reported the results of a meta-analysis on the use of 18 F-FDG PET/ CT for radiotherapy dose planning. According to this meta-analysis, a change in target definition was 36% in patients with a former staging PET, and 43% and 26% in patients without a staging PET for NSCLC and SCLC, respectively. The corresponding summary estimates of a change in treatment intent from curative to palliative treatment were 20% and 22% and 9%, respectively.
Another recent meta-analysis demonstrated that functional lung imaging, including PET, may have potential utility in radiation therapy planning and delivery [34].

Diagnosis of Lung Cancer Recurrence
Although there are no conclusive data to support the survival benefits of early detection or early treatment for recurrence of lung cancer, an early and accurate diagnosis of recurrence is critical to optimize therapy. A meta-analysis [35] was performed to assess the diagnostic value of 18 F-FDG PET and PET/CT for cases of recurrent lung cancer. In the patient-based analysis performed, 18 F-FDG PET and PET/CT were found to provide better detection of lung cancer recurrence compared to CT. Indeed, the pooled sensitivity for 18 F-FDG PET, PET/CT, and CT were 94% (95%CI: 91-97%), 90% (95%CI: 84-95), and 78% (95%CI: 71-84%), respectively while the pooled specificity for 18 F-FDG PET, PET/CT, and CT were 84% (95%CI: 77-89%), 90% (95%CI: 87-93%), and 80% (95%CI: 75-84%), respectively.

Prognostic Evaluation in NSCLC
In their meta-analysis, Paesmans et al. [36] assessed the prognostic value of primary tumour maximum standardized uptake value (SUVmax) at 18 F-FDG PET for overall survival (OS) of NSCLC patients. At multivariate analysis, SUVmax was found to be independently associated with survival. The hazard ratio (HR) for SUVmax was 1.58 (95%CI: 1.27-1.96). Despite the SUVmax represents the most widely applied semi-quantitative PET parameter in clinical practice, volumetric PET parameters, including metabolic tumour volume (MTV) and total lesion glycolysis (TLG), have been also used to reflect disease burden and tumour aggressiveness in NSCLC. A first meta-analysis performed by Liu et al. [37] explored the prognostic value of SUVmax, MTV, and TLG on diseasefree survival (DFS) and OS in surgical NSCLC patients. The pooled HRs for OS were 1.52 for SUVmax, 1.91 for MTV, and 1.94 for TLG. On the basis of these results, the authors stated that high values of SUVmax, MTV, and TLG are able to predict a higher risk of recurrence or death in patients with surgical NSCLC, suggesting the use of 18 F-FDG PET/CT to select patients who are at high risk of disease recurrence or death as the best candidates from aggressive treatments. Other authors [38] conducted a meta-analysis on the prognostic value of MTV and TLG in NSCLC patients. A worse prognosis was observed in patients with high MTV (HR: 2.31) and with high TLG (HR: 2.43).
Han and colleagues [39] performed a metaanalysis exploring prognostic value of texture parameters derived by 18 F-FDG PET in patients with lung cancer. They concluded that there is insufficient evidence to support the prognostic value of texture analysis in 18 F-FDG PET in lung cancer.
Another interesting application of 18 F-FDG PET is the ability to predict long-term results after radiation therapy. Dong and co-workers [40] explored the prognostic relevance of SUVmax at 18 F-FDG PET for early-stage NSCLC patients receiving stereotactic body radiation therapy (SBRT). The authors found that those NSCLC patients presenting with high levels of pre-SBRT SUVmax had poorer OS and local control and higher risk of distant metastases. These findings were confirmed by another metaanalysis [41] showing that both pre-radiotherapy and post-radiotherapy primary tumour SUVmax can predict the outcome of patients with NSCLC treated with radiotherapy.
Other authors [42] have summarized the prognostic value of early response at 18 F-FDG PET in NSCLC patients treated with tyrosine-kinase inhibitors (TKI). Early response of patients with NSCLC treated with TKIs identified on 18 F-FDG PET was found to be associated with improved OS and progression-free survival (PFS).

Management of SCLC
The role of 18 F-FDG PET in the management of SCLC has been largely investigated in the last decades. A systematic review and meta-analysis performed by Lu et al. [43] to evaluate the diagnostic accuracy of 18 F-FDG PET/CT in the pretherapeutic staging of patients with SCLC demonstrated a pooled sensitivity and specificity of 97.5% (95%CI: 94.2-99.2%) and 98.2% (95%CI: 94.9-99.6%), respectively, for the detection of extensive disease in SCLC patients. Therefore, evidence-based data suggest the role of 18 F-FDG PET/CT for discriminating between limited and extensive disease in SCLC.
The prognostic value of the SUVmax of primary SCLC at 18 F-FDG PET was recently investigated through a meta-analytic study [44]: the pooled HR for OS was 1.13 (95%CI: 1.05-1.22), thus indicating that SCLC patients with high SUVmax may have poorer prognosis.

Evidence-Based Data on PET in Pleural Tumours
Three meta-analyses assessed the role of 18 F-FDG PET or PET/CT in the characterization of pleural lesions [45][46][47], whereas meta-analyses on the role of 18 F-FDG PET/CT in staging, restaging, prognostic or treatment response evaluation of pleural tumours are currently lacking. 18 F-FDG-PET and PET/CT demonstrated to be accurate diagnostic imaging methods in the differential diagnosis between malignant and benign pleural lesions in patients with or without known cancer; nevertheless, possible sources of falsenegative and false-positive results should be kept in mind [45,46]. In patients without known cancer, sensitivity and specificity of 18 F-FDG-PET and PET/CT were 95% (95%CI: 92-97%) and 82% (95%CI: 76-88%), respectively [45]. In patients with known cancer, pooled sensitivity was 86% (95%CI: 80-91%) and pooled specificity was 80% (95%CI: 73-85%) [46]. Porcel et al. in their meta-analysis [47] demonstrated that semi-quantitative PET assessment had a significantly lower sensitivity for diagnosing malignant pleural effusions than visual assessments. The pooled sensitivity and specificity of 18 F-FDG PET/CT using semi-quantitative interpretation for identifying malignant pleural effusions were 81% and 74%, respectively. The moderate accuracy of semi-quantitative PET assessment precludes its routine recommendation for discriminating malignant from benign pleural effusions.
Notably, meta-analyses on the role of 18 F-FDG PET/CT in staging, restaging, prognostic or treatment response evaluation of TETs are currently lacking.

5.5
Evidence-Based Data on PET in Esophageal Tumours

Staging
The real and unquestionable additional diagnostic value of 18 F-FDG PET/CT in comparison to conventional imaging methods is in evaluating distant metastases (M staging) of esophageal cancer [49], whereas recent evidence-based articles have addressed the performance of 18 F-FDG PET/CT for detecting lymph nodal metastases (N staging). Jiang et al. [50] found that the pooled sensitivity and specificity estimates of 18 F-FDG PET/CT for detecting regional lymph nodal metastases at staging were 66% (95%CI: 51-78%) and 96% (95%CI: 92-98%), respectively. The corresponding values on a per-patient analysis were 65% (95% CI: 49-78%) and 81% (95%CI: 69-89%), respectively. Overall, 18 F-FDG PET/CT has a moderate to low sensitivity and a high to moderate specificity for detection of regional nodal metastases in esophageal cancer. Therefore, extending the extent of lymph node dissection or radiotherapy target volume is necessary after the diagnosis of regional nodal metastases by 18 In another meta-analysis [51], Hu et al. evaluated the diagnostic performance of 18 F-FDG PET/ CT for the assessment of preoperative lymph node metastases in patients with esophageal cancer. In patients without neoadjuvant treatment, 18 F-FDG PET/CT had a pooled sensitivity and specificity of 57% (95%CI: 45-69%) and 91% (95%CI: 85-95), respectively. In patients who received neoadjuvant treatment, 18 F-FDG PET/ CT had a pooled sensitivity and specificity of 53% (95%CI: 35-70%) and 96% (95%CI: 86-99%), respectively. Therefore, 18 F-FDG PET/CT has a high specificity but a low sensitivity; thus, it cannot accurately detect the lymph nodal involvement in patients with esophageal cancer.
In this setting, cervical ultrasonography has very limited additional diagnostic value as supplement to a negative 18 F-FDG PET/CT in the detection of cervical lymph node metastases during the initial staging of patients with esophageal cancer, as demonstrated by Goense et al. [53].

Restaging
Restaging after neoadjuvant therapy aims to reduce the number of patients undergoing oesophagectomy in case of distant (interval) metastases. Kroese et al. [54] assessed the diagnostic performance of 18 F-FDG PET or PET/CT for the detection of distant interval metastases after neoadjuvant therapy in patients with esophageal cancer. The pooled proportion of patients in whom true distant interval metastases were detected by 18 F-FDG PET or PET/CT at restaging was 8% (95%CI: 5-13%). The pooled proportion of patients in whom false-positive distant findings were detected by 18 F-FDG PET or PET/CT at restaging was 5% (95%CI: 3-9%). In conclusion, 18 F-FDG PET or PET/CT at restaging after neoadjuvant therapy for esophageal cancer can considerably impact on treatment decisionmaking. However, pathological confirmation of suspected lesions is needed.
Cong et al. [55] assessed the value of 18 F-FDG PET or PET/CT for response prediction of primary tumour in patients with esophageal cancer during (group A) or after (group B) neoadjuvant chemoradiotherapy. The pooled sensitivity and specificity were 85% (95%CI: 76-91%) and 59% (95%CI: 48-69%), respectively, in group A. The equivalent values were 67% (95%CI: 60-73%) and 69% (95%CI: 63-74%), respectively, in group B. Interestingly, the pooled sensitivity was 90% in the studies that enrolled patients with esophageal squamous cell carcinoma merely in group B. According to the present data, 18 F-FDG PET/ CT should not be used routinely to guide treatment strategy in esophageal cancer patients, but an additional value is expected in patients with esophageal squamous cell carcinoma treated with neoadjuvant chemoradiotherapy.
Goense et al. [56] assessed the diagnostic performance of 18 F-FDG PET or PET/CT for diagnosing recurrent esophageal cancer after initial treatment with curative intent. Pooled estimates of sensitivity and specificity for 18 F-FDG PET and PET/CT in this setting were 96% (95%CI: 93-97%) and 78% (95%CI: 66-86%), respec-tively. Therefore, 18 F-FDG PET and PET/CT are reliable imaging modalities with a high sensitivity and moderate specificity for detecting recurrent esophageal cancer after treatment with curative intent. However, histopathologic confirmation of PET/CT-suspected lesions is required, because a considerable false-positive rate is noticed.

Predictive and Prognostic Value
Han et al. [57] performed a meta-analysis on the prognostic value of volumetric parameters (MTV and TLG) derived from pretreatment 18 [58] performed a metaanalysis on the predictive value of 18 F-FDG PET for assessing DFS and OS in esophageal and oesophagogastric junction cancer after neoadjuvant chemoradiation therapy. The pooled HRs for complete metabolic response versus no response were 0.51 for OS (95%CI: 0.4-0.64) and 0.47 for DFS (95%CI: 0.38-0.57), respectively. Therefore, metabolic response on 18 F-FDG PET is a significant predictor of long-term survival.
Lastly, Zhu et al. [59] performed a metaanalysis on the prognostic significance of SUVmax on 18 F-FDG PET/CT in patients with localized oesophagogastric junction cancer receiving neoadjuvant chemotherapy/chemoradiation therapy. Significant prognostic values of SUVmax before and during therapy in localized oesophagogastric junction cancer were not found. Conversely, relative changes in 18 F-FDG-uptake after therapy are significant prognostic markers for OS and DFS.