evaluated for molecular biomarkers as more and more therapies are being developed to treat cancers which harbor these speci c mutations. The National Comprehensive Cancer Network (NCCN) guidelines on NSCLC recommends testing for ALK rearrangements, BRAF mutations, epidermal growth factor receptor (EGFR) mutations, METex14 skipping mutations, neurotrophic tyrosine receptor kinase 1/2/3 (NTRK1/2/3) gene fusions, rearranged during transfection (RET) rearrangements, and c-ros oncogene 1 (ROS1) rearragements along with immunohistochemical testing for programmed death ligand 1 (PD-L1) [7]. This list will likely increase quickly over the next decade as new therapeutic targets are identi ed.

Initial evaluation of a patient with suspected lung cancer begins with a thorough history and physical examination. Asymptomatic patients will commonly present after an incidental nding on chest imaging or dedicated lung cancer screening. Every effort should be made to review prior images to help determine the age and growth pattern of said abnormalities. Intra-thoracic effects of lung cancer include a wide range of symptoms including cough, dyspnea, and weight loss [8]. When evaluating for extra thoracic metastasis, the most common sites in descending order of frequency are nervous system, bone, liver, lung, and adrenal glands [9]. Extra-thoracic effects of metastatic disease depend on the organ/system involved. For example, a patient with suspected lung cancer who is experiencing new onset headaches may warrant a brain computerized tomography (CT) or magnetic resonance imaging

(MRI) to search for metastatic disease. Paraneoplastic syndromes can involve various systems that include dermatologic, rheumatologic, neurologic, endocrine, hematologic, renal, and ophthalmologic systems [10]. Paraneoplastic syndromes can occur in up to 10% of patients with lung cancer; the two most common being humoral hypercalcemia of malignancy (HHM) in squamous cell carcinoma (SCC) and the syndrome of inappropriate antidiuretic hormone secretion (SIADH) in small cell lung cancer [11].

This chapter will be presented in a case-based format that highlights the proper initial evaluation, diagnosis, and staging of lung cancer. The cases will highlight the important clinical aspects for clinicians to consider when evaluating these patients.

Case 1

A 55-year-old man with a 40 pack-year smoking history came to clinic with complaints of shoulder pain. During evaluation he underwent a CT chest (Fig. 28.2a) which revealed a large solid appearing mass in the left upper lobe measuring 8.4 × 7.4 cm abutting the mediastinum and chest wall. There were enlarged left hilar and mediastinal lymph nodes with 4 L measuring 1.2 cm. Lung parenchyma revealed widespread multilobular emphysema. Skeletal exam was unremarkable.

The purpose of searching for extrathoracic disease in NSCLC is to detect metastatic disease at common metastatic sites, such as the adrenal

Fig. 28.2  (a) Large 8.5 × 7.9 cm left upper lobe mass abutting and invading the superior mediastinum. (b) Large hypermetabolic left apical mass with a max SUV of 28.9.

(c) Hypermetabolic right adrenal mass measuring 3.6 cm with a max SUV of 17.5

glands, liver, brain, and skeletal system. A biopsy of a distant site of metastasis would both diagnose and stage, thereby sparing the patient futile surgical interventions. The current practice is based on clinical evaluation of organ-speci c symptoms, constitutional signs and symptoms, along with simple laboratory tests to aid in workup of extrathoracic disease. If patients have abnormal symptoms, physical exam, or blood tests, there is a high likelihood of metastasis [12]. One study found that distant metastases may become evident as early as 4 weeks in 3% of untreated patients and may increase to 13% as early as 8 weeks. This leads the authors to propose complete restaging after a 4 to 8 weeks delay in therapy [12]. It is also known that patients with an oligometastatic lesion in a single organ such as brain, liver, bone, distant lymph node, skin, peritoneum, or adrenal gland have better survival than those with multiple extrathoracic lesions. These patients may be candidates for surgical resection or local ablative therapy with curative intent [13]. Five-year relative survival rate of patient with distant spread of lung cancer is around 7% as compared to 60% in localized disease [14]. The median overall survival of patients with stage IV lung cancer ranges between 7.0 and 12.2 months depending on treatment, histologic type, and other associated factors such as patient co-morbidities [14]. One study found that survival for those with advanced stage disease, those who have a targetable mutation or are responsive to immunotherapy, can be signi­ cantly longer, 20 months vs. 12.2 months [15]. Given the impact of extrathoracic disease on patient prognosis, it is crucial to choose the most appropriate imaging and biopsy modalities. Histopathologic con rmation of the extrathoracic disease is necessary to confer the highest and most accurate stage to the patient. In addition, it is incumbent on the pulmonologist to provide a “molecular stage” which may provide signi - cantly better treatment options for patients.

Computerized tomography of the chest, CT or MRI with contrast of the brain, and 99mTc nuclear imaging of the skeletal system are the conventional staging studies; however, data over the last decade con rm the superiority of the per-

formance characteristics of positron emission tomography (PET) and PET-CT scans, compared with conventional scans, in the evaluation of metastatic disease in key speci c distant sites. Speci cally, PET scan reveals unsuspected metastases in 6–37% of patients [12]. In the results of a 2013 meta-analysis of nine studies, fuorodeoxyglucose (FDG) PET-CT had a sensitivity of 93%, a speci city of 96%, a positive likelihood ratio of 28.4%, and a negative likelihood ratio of 0.08% for detection of distant metastases [16]. Additionally, PET is cost-­ effective compared with CT. Søgaard et al. reported that PET-CT increased the cost by 3927 Euros and that 5 PET-CT scans are needed to prevent one noncurative surgical resection [17]. The American College of Chest Physicians (ACCP) guidelines recommends PET to evaluate for extrathoracic metastasis, except for brain metastasis (where PET is not useful), in patients with a normal clinical evaluation and no suspicious extrathoracic abnormalities on chest CT being considered for curative-intent treatment [12]. Advanced thoracic lesions and mediastinal lymphadenopathy, particularly N2 disease, are associated with higher rates of asymptomatic metastases and in those cases PET should be performed [18].

Important limitations relating to false positive and false negative scans exist within extrathoracic disease imaging and a positive PET scan requires careful clinical correlation with biopsy con rmation to accurately stage the patient [12]. Clinicians must have pathologic evidence of metastatic disease that was suggested by PET unless there are overwhelming ndings on the scan such as multiple bilateral nodules or multiple sites of uptake in the bony skeleton.

Pleura and Pleural Efusion

There are limited data to suggest that CT and PET scan are useful in identifying malignant pleural effusion (MPE); although, much of the data pertain to non-pulmonary malignancies [12]. Certain CT features such as pleural thickening, pleural nodularity, or lung parenchymal

lesions in close proximity to the pleura heighten suspicion for pleural metastases. One study found the sensitivity of CT for MPE can be increased by its integration with FDG-PET, reaching 93% in comparison to 70% with CT alone [19]. No radiological study can substitute for cytohistological con rmation if pleural malignancy is a consideration.

Adrenal and Hepatic Metastases

Incidental adrenal nodules are found in 20% of patients with NSCLC [20]. Although it is common to encounter adrenal masses in a routine CT scan, unilateral adrenal masses measuring greater than 3 cm in a patient with suspected lung cancer is more likely to be a metastasis rather than a benign lesion [20].

Computerized tomography, MRI, PET, percutaneous biopsy, and even adrenalectomy can be used to help distinguish benign from malignant disease [12]. Delayed contrast-enhanced CT can help aid the differentiation of benign and malignant lesions, but is rarely performed. MRI has not shown superiority when compared to CT because of the considerable overlap of signal intensity in benign and malignant lesions. PET scan outperforms both CT and MRI in identifying malignant lesions of the adrenal gland. A recent meta-­analysis of nine studies evaluating the diagnostic accuracy of PET-CT for the detection of adrenal metastasis noted a pooled sensitivity of 89% and a speci city of 90%. False-negative results can occur in metastases with hemorrhage, necrosis, and in lesions measuring less than 1 cm. Adrenal hyperplasia, adrenal adenoma, and infections can result in false-positive results [21].

Most liver lesions are benign cysts or hemangiomas, but a contrast CT scan or ultrasound is often required to establish a likely diagnosis. Although there are limited data with NSCLC, PET-CT and MRI have been used to detect liver metastases. The PET scan has an accuracy of 92–100% with rare false positives results [12]. A recent meta-analysis showed that MRI has a higher sensitivity and speci city when compared

to CT scan in detecting liver lesions (93.1% vs. 82.1% for sensitivity, 87.3% vs. 73.5% for speci-city) [22].

Brain

Magnetic resonance imaging is the gold standard to evaluate for brain metastases. ACCP guidelines suggest routine MRI brain for clinical stage III or IV NSCLC, but not for asymptomatic stage I or II disease [23]. CT scans are also used to evaluate metastasis to the brain. Because of brain abscesses, gliomas, and other lesions, CT scans under perform in detecting brain metastasis. The false negative and false positive rates of cranial CT are reported to be 3% and 11%, respectively [12]. Detection of brain metastases is a problem for PET-CT because the high background brain FDG uptake can mask the small size of most brain metastases. Lesions can be either hypermetabolic or hypometabolic (Fig. 28.3) [24]. The results of a meta-analysis showed pooled sensitivities of 21% and 77% and speci cities of 100% and 99% for PET and MRI, respectively [25].

Bone

Radionucleotide bone scintigraphy can be used to detect bone metastases but is troubled with a high false positive rate owing to the frequency of degenerative and traumatic skeletal damage. PET appears to have excellent performance characteristics in assessing bone metastases. One meta-­ analysis of patients with lung cancer revealed that FDG PET-CT was more accurate for the diagnosis of bone metastases when compared to MRI or bone scintigraphy. In addition, combined PET-CT has better performance characteristics than any other method for detecting bony metastases [26].

Case 1 Continued

The patient underwent a PET-CT which demonstrated a hypermetabolic 8.5 cm left apical mass along with increased uptake in left hilum, medi-

astinum, and thoracic vertebrae. There was a hypermetabolic left pleural nodule and a 3.6 cm right adrenal lesion (Fig. 28.2b, c).

Once imaging is suspicious for extrathoracic spread, the next step is tissue acquisition. The goal of any biopsy is to provide adequate samples for the pathologist to arrive at a de nitive histopathologic diagnosis and have adequate tissue for molecular analysis to help dictate therapy.

Although bone biopsy has a high rate of success for diagnosing metastatic disease, it often has inadequate tissue for molecular analysis. VanderLaan et al. showed that of the 207 patients with concurrent testing, the failure rate for bone-­ derived specimens were 23.1% for EGFR, 15.4% for KRAS, and 23.1% for ALK. This is attributed to the decalci cation process of the tumor tissue which alters the DNA leading to failure of molecular analysis [27]. Adrenal lesions with radiographic features suspicious for malignancy can be sampled by percutaneous ne needle aspiration/biopsy or, less commonly, by endoscopic ultrasoundne needle aspiration (EUSFNA) [12]. Brain biopsy is not routinely performed but should be considered when the diagnosis of brain metastases is in doubt. This is particularly important in patients with a concern for oligometastatic disease. Tissue sampling of a distant metastatic site is not necessary if there is overwhelming radiographic evidence of metastatic disease in multiple sites [12]. Patients with suspected lung cancer who present with a pleural

effusion should undergo ultrasound-guided thoracentesis with a goal of drawing at least 50 mL of pleural fuid. The diagnostic yield for pleural fuid cytology has a reported mean sensitivity of 72%, with a range of 49–91%. The sensitivity of pleural fuid cytology increases a further 27% with a second thoracentesis, but only a 5% yield from a third. If cytology after two subsequent thoracentesis is negative for cancer, one should consider imaging guided biopsy, surgical biopsy or thoracoscopy [28].

Biomarkers

Historically, patients with advanced NSCLC received cytotoxic chemotherapy regimens, including platinum-based chemotherapies. The 5-year relative survival for metastatic disease is approximately 7% in these individuals [14]. Sensitizing mutations in EGFR, with regard to NSCLC, were rst described in 2004 [29]. Nearly two decades later, there have been incredible advancements in detecting and treating lung cancer patients with approved treatments for which there are speci c mutations found on the tumor. Why is this important? We know that patients who receive matched targeted therapy have superior overall survival compared with patients who receive cytotoxic chemotherapy or no therapy at all. Gutierres et al. reported survival data for 805 patients with advanced NSCLC. The 131 patients