Lung cancer is by far the most prevalent malignancy in the United States, with approximately 170,000 new cases being identified each year and 157,000 deaths. Detection, diagnosis, staging, and monitoring of the disease therefore are a significant part of thoracic radiology. New techniques are changing the way these patients are examined.

The Solitary Nodule: What Is It?

As interest in screening for lung cancer grows in popularity (see boxes, pages 36 and 40), an increasingly common challenge to thoracic imaging is characterization of the solitary nodules that show up on so many chest images. Is a nodule benign? Between 30% and 50% are. Or is it malignant? Can biopsy be avoided?

Positron emission tomography (PET) with fluorine 18-labeled deoxyglucose (FDG), with its ability to reveal areas of high metabolic activity, has demonstrated clinical utility and cost-effectiveness in nodule characterization and is reimbursed by the Centers for Medicare and Medicaid Services.1-3 However, both false-positive and false-negative results can create problems. Granulomatous disease can cause false-positive scans, and smaller nodules may not concentrate enough isotope to be visible. Helical CT has become a standard characterization method because it eliminates the overlying tissue and produces greater contrast between the lesion and the lung tissue than is possible with the standard chest radiograph. Of even greater utility is multidetector array CT, the speed of which largely eliminates ventilatory and cardiac motion that obscure details. These images permit analysis of lesion edge, shape, and internal features; size (including volume); and growth rate,4 which often permits characterization. For example, spiculated lesions are nearly always malignant (although lesions with smooth edges are not necessarily benign), and pseudocavitation likewise strongly suggests malignancy.

Computer-aided detection may further reduce the number of patients who require biopsy. The Rossmann Laboratories at the University of Chicago have demonstrated an artificial neural network to determine whether a pulmonary nodule is benign or malignant. In an early trial, the system, working with digitized chest radiographs, proved more accurate than radiologists.5

Considerable research also is being directed at exploiting chemical differences between lung cancers and other causes of pulmonary nodules. One approach is single-photon emission computed tomography (SPECT) to capture signal from a radiolabeled antitumor antibody. Early trials suggest that this method may prove more cost-effective than PET.6 Alternatively, it may be possible to exploit the tendency of lung cancers to overexpress somatostatin receptor by imaging a 99mtechnetium-labeled somatostatin analog. One line of research is aimed at creating ligands with particularly high affinity for the receptor subtypes most common on lung cancers.7 A team at the University of Ulm in Germany has shown that PET using fluorine 18-labeled deoxyfluorothymidine to image cell division may help identify cancer in a nodule, as the cells have faster reproductive rates than do most benign tissues.8

Not all cancer in the lungs is primary there. As noted by Bradley Sabloff, MD, of the Division of Diagnostic Imaging at The University of TexasMD Anderson Cancer Center, Houston,  at the September conference on Oncologic Imaging for the Practicing Radiologist in San Antonio, “metastatic disease is the most common chest malignancy, and the chest acquires more metastases than any other system.” At autopsy, 20% to 54% of cancer patients have lung metastases, most often from tumors of the breast, colon, melanoma, head and neck, and kidney. Spiral, especially high-resolution, CT is the gold standard for evaluating cancer patients for thoracic metastases. Four-slice CT with 7.5-mm slices and 3.75-mm slice reconstruction with standard and lung algorithms are the most useful, according to Sabloff. Images are viewed on a workstation in the ciné mode.

Staging of Lung Cancer

Once a lung cancer is confirmed, the next question is whether it has spread locally, regionally, or to distant sites; that is, whether the tumor is resectable. Computed tomography is superior to chest radiography in detecting chest wall invasion and mediastinal, diaphragmatic, and nodal involvement and thus can help the surgeon design the operative approach and direct mediastinoscopy for any necessary biopsy. Two common sites of metastases, namely the liver and the adrenal glands, can be imaged in the same scan. Nodal size is the only criterion of abnormality; unfortunately, metastases can be present in normal-size nodes, and large nodes do not always contain tumor. There is growing enthusiasm for FDG-PET, which “has consistently proved superior to structure-based imaging modalities in both the diagnosis and staging of lung cancer.”9 The Stroobants et al study showed FDG-PET had an accuracy of 81% to 96%  in staging the chest; it was especially useful in evaluating nodes of normal size and in differentiating hyperplasia from cancer and, like CT, can suggest appropriate sites for biopsy. Some authors have suggested that a patient with a negative PET scan does not require mediastinoscopy. If a CT/PET scanner is available, with its combined functional and anatomic images, it can be used to reduce the likelihood that a patient with unresectable disease will be operated on.

A further use of FDG-PET is in searching for distant metastases, although there is some debate about which patients are appropriate candidates (see below). Whole-body PET permits assessment for intrathoracic disease and extrathoracic disease in a single study.10 The study is 100% sensitive and 80% specific for adrenal metastases, which are established in as many as 20% of patients at presentation. If the scan is negative, the lesion is assumed to be benign; if it is positive, a biopsy is obtained. A PET scan finds occult extrathoracic metastases (ie, unresectable disease) in 11% to 14% of patients thought to have resectable disease and alters the treatment plan in as many as 41% of patients.10

Routine brain imaging by any modality is not cost-effective, as most patients with brain metastases have signs and symptoms.  Similarly, skeletal metastases usually produce laboratory abnormalities or symptoms, although 99mTc-MDP scintigraphy may be required in a few patients. A team at Duke University, which recently completed a comprehensive review of published data, suggested that routine imaging of asymptomatic patients for distant metastases may not be necessary.11

Follow-Up of Treatment

Many consider PET more accurate than traditional methods in determining the response of a lung cancer to chemotherapy or radiation.9 But imaging has a role in follow-up beyond assessment of treatment response or searching for recurrences.12 It also provides information on complications. Lung injury is a common source of morbidity and mortality in patients given cancer chemotherapy.13 Pulmonary toxicity may appear after the first dose of drugs, often as a hypersensitivity-type reaction, or be delayed and evidenced as nonspecific interstitial pneumonitis or bronchiolitis obliterans organizing pneumonia. High-resolution CT is the classic method of identifying such toxicity. The damage often regresses if the drugs are stopped or steroids are given early in the disease course, sparing the patient potentially fatal lung injury.

Conclusion

The new capabilities of imaging have other applications not touched on here. For example, virtual bronchoscopy has been as impressive as other methods of noninvasive endoscopy, and endobronchial ultrasonography has excellent sensitivity as a staging method when used by experienced operators.14 Magnetic resonance imaging, traditionally used only to answer questions raised by CT or in patients with contrast allergy, may find a greater role in diagnosis, staging, and follow-up as new sequences increase its yield of functional as well as anatomic information.15 As the number of lung cancer patients continues to rise, the pressure on radiologists to do even more with their modalities will increase.

Judith Gunn Bronson, MS, is a contributing writer for Decisions in Axis Imaging News.

References:

  1. Shon IH, O’Doherty MJ, Maisey MN. Positron emission tomography in lung cancer. Semin Nucl Med. 2002;32:240—271.
  2. Czernin J, Phelps ME. Positron emission tomography scanning: current and future applications. Annu Rev Med. 2002;53:89—112.
  3. Lowe VJ. Positron emission tomography scanning in lung cancer. Respir Care Clin N Am. 2003;9:119—136.
  4. Ko JP, Naidich DP. Lung nodule detection and characterization with multislice CT. Radiol Clin N Am. 2003;41:575—597.
  5. Nakamura K, Yoshida H, Engelmann R, et al. Computerized analysis of the likelihood of malignancy in solitary pulmonary nodules with use of artificial neural networks. Radiology. 2000;214:823—830.
  6. Machac J, Krynyckyi B, Kim C. Peptide and antibody imaging in lung cancer. Semin Nucl Med. 2002;32:276—292.
  7. Blum J, Handmaker H, Rinne NA. Technetium labeled small peptide radiopharmaceuticals in the identification of lung cancer. Curr Pharm Des. 2002;8:1827—1836.
  8. Buck AK, Halter G, Schirrmeister H, et al. Imaging proliferation in lung tumors with PET: 18F-FLT versus 18F-FDG [see comment]. J Nucl Med. 2003;44:1426—1431.
  9. Stroobants S, Verschakelen J, Vansteenkiste J. Value of FDG-PET in the management of non-small cell lung cancer. Eur J Radiol. 2003;45:49—59.
  10. Higashi K, Matsunari I, Ueda Y, et al. Value of whole-body FDG PET in management of lung cancer. Ann Nucl Med. 2003;17:1—14.
  11. Toloza EM, Harpole L, McCrory DC. Noninvasive staging of non-small cell lung cancer: a review of the current evidence. Chest. 2003;123(1 suppl):137S—146S.
  12. Colice GL, Rubins J, Unger M, American College of Chest Physicians. Follow-up and surveillance of the lung cancer patient following curative-intent therapy. Chest. 2003;123(1 suppl):272S—283S.
  13. Erasmus JJ, McAdams HP, Rossi SE. Drug-induced lung injury. Semin Roentgenol. 2002;37:72—81.
  14. van Zandwijk N. New methods for early diagnosis of lung cancer. Lung Cancer. 2002;38:S9—S11.
  15. Ohno Y, Sugimura K, Hatabu H. MR imaging of lung cancer. Eur J Radiol. 2002;44:172—181.