HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) All Versions of this Article: 2373040966v1 237/3/803    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Munden, R. F. Articles by Stewart, D. J. Search for Related Content PubMed PubMed Citation Articles by Munden, R. F. Articles by Stewart, D. J.

Imaging of the Patient with Non–Small Cell Lung Cancer¹

¹ From the Division of Diagnostic Imaging, Department of Diagnostic Radiology (R.F.M.), Department of Thoracic and Cardiovascular Surgery (S.S.S.), Division of Radiation Oncology (C.W.S.), and Division of Cancer Medicine (D.J.S.), University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Blvd, Houston, TX 77030. Received June 2, 2004; revision requested August 12; revision received October 27; accepted December 15. Address correspondence to R.F.M. (e-mail: rmunden{at}di.mdacc.tmc.edu).

	ABSTRACT

TOP ABSTRACT INTRODUCTION SURGEON'S PERSPECTIVE RADIATION ONCOLOGIST'S... MEDICAL ONCOLOGIST'S PERSPECTIVE RADIOLOGIST IN MULTIDISCIPLINARY... CONCLUSION References

 
Lung cancer is the most common type of cancer and is the leading cause of cancer deaths in the United States for both men and women. Even though the 5-year survival rate of patients with lung cancer remains dismal at 14% for all cancer stages, treatments are improving and newer agents for lung cancer appear promising. Therefore, an accurate assessment of the extent of disease is critical to determine whether the patient is treated with surgical resection, radiation therapy, chemotherapy, or a combination of these modalities. Radiologic imaging plays an important role in the staging evaluation of the patient; however, radiologists need to be aware that there are also important differences in what each specialist needs from imaging to provide appropriate treatment. This article reviews the role of imaging in patients with non–small cell lung cancer, with an emphasis on the radiologic imaging findings relevant for each specialty.

© RSNA, 2005

	INTRODUCTION

TOP ABSTRACT INTRODUCTION SURGEON'S PERSPECTIVE RADIATION ONCOLOGIST'S... MEDICAL ONCOLOGIST'S PERSPECTIVE RADIOLOGIST IN MULTIDISCIPLINARY... CONCLUSION References

 
Lung cancer is the most common type of cancer and is the leading cause of cancer deaths in the United States for both men and women. In 2002, it was estimated that 169 500 cases of lung cancer would be diagnosed and that approximately 154 400 deaths (1) would occur. The problem continues to increase, with an estimated 173 770 new cases and 160 440 deaths in 2004 (2). Despite heroic efforts, the overall 5-year survival rate in patients with lung cancer remains dismal at 14% for all stages (clinical staging): 61% for stage IA, 38% for stage IB, 34% for stage IIA, 24% for stage IIB, 13% for stage IIIA, 5% for stage IIIB, and 1% for stage IV (3).

Staging of newly diagnosed non–small cell lung cancer (NSCLC) is performed according to the International System for Staging Lung Cancer. This system describes the extent of NSCLC in terms of the size, location, and extent of the primary tumor (T descriptor), the presence and location of lymph node involvement (N descriptor), and the presence or absence of distant metastatic disease (M descriptor) (see table 1 in reference 3). Because the extent of the disease determines whether the patient is treated by means of surgical resection, radiation therapy, chemotherapy, or a combination of these modalities, radiologic imaging plays an important role in the staging evaluation (4,5). Although accurate staging is critical regardless of the treatment method used, there are important differences in what each specialist needs from imaging to provide appropriate treatment. This article reviews the role of imaging in patients with NSCLC, with an emphasis on the radiologic findings relevant for each specialty.

	SURGEON'S PERSPECTIVE

TOP ABSTRACT INTRODUCTION SURGEON'S PERSPECTIVE RADIATION ONCOLOGIST'S... MEDICAL ONCOLOGIST'S PERSPECTIVE RADIOLOGIST IN MULTIDISCIPLINARY... CONCLUSION References

 
Preoperative Imaging
Because surgical resection of lung cancer offers the best chance of cure, accurate staging is crucial for the selection of patients for surgery. In general, patients with good performance status and clinical stage I or II disease, as well as some patients with stage IIIA disease, are considered for surgery, whereas most patients with stage IIIB (except those with T4N0) and all of those with stage IV disease are not considered (6). This conclusion is based on pathologic survival data that show that patients with pathologic stage I have a 5-year survival rate of 67% for stage IA, 57% for IB, 55% for IIA, 39% for IIB, and 23% for IIIA (3). This is in comparison with results of clinical staging that show a survival rate of 61% for stage IA, 38% for stage IB, 34% for stage IIA, 24% for stage IIB, 13% for stage IIIA, 5% for stage IIIB, and 1% for stage IV (3). Because pathologic staging is not available prior to treatment, the method of treatment is based on clinical staging, and diagnostic imaging is a critical component in the evaluation of the primary tumor (T descriptor), lymph nodes (N descriptor), and metastatic disease (M descriptor). The common imaging modalities used for staging in patients with NSCLC include chest radiography, computed tomography (CT), magnetic resonance (MR) imaging, positron emission tomography (PET), and fused PET/CT.

While many lung cancers are diagnosed on chest radiographs after the patient presents with symptoms, the standard posteroanterior radiograph is of limited use for staging. Even if obvious bone metastasis is evident on chest radiographs, patients still need to undergo further imaging to establish the extent of the disease so that the effectiveness of treatment can be monitored. CT of the chest is the most common radiologic study performed after chest radiography to evaluate lung cancer in a patient (7). Complementary to CT is imaging with PET by using fluorine 18 (¹⁸F) fluorodeoxyglucose (FDG). PET is increasingly being used for diagnosis and staging in patients with lung cancer because it has been shown to enable more accurate assessment for extrathoracic metastases in these patients (8). However, radiologists should be aware that bronchioloalveolar cell carcinoma is often not FDG avid and a false-negative rate of 40% has been reported in PET studies of patients with bronchioloalveolar cell carcinoma (9).

An understanding of the primary tumor (T descriptor) is necessary for staging, as well as for the surgical treatment plan, and, therefore, it is important for radiologists to appropriately describe the primary tumor in their reports.

T1 tumors are technically the easiest to resect because they are smaller than 3 cm in size and are limited to the lung parenchyma. An important finding of T1 tumors for radiologists to note is whether the tumor is near the main pulmonary artery, because pneumonectomy or pulmonary arterioplasty may be indicated (10). It is also important for the radiologist to differentiate a tumor that crosses an incomplete fissure from one that truly involves a fissure. Fissures are composed of two layers of visceral pleura from adjacent lobes. Complete fissures extend from the peripheral pleural surface to the hilum, whereas incomplete fissures do not extend completely to the hilum. A tumor medial to an incomplete fissure resides completely within lung parenchyma and, if smaller than 3 cm, is classified as a T1 tumor, whereas if it involves the pleura it would be a classified as a T3 tumor. Although technically such a T1 tumor does not touch the pleura, it is referred to as crossing an incomplete fissure because it lies in the projected pathway of the incomplete fissure if it were a complete fissure. The importance of noting that a tumor crosses an incomplete fissure is to indicate that it does not involve the pleura, it resides in two lobes of the lung, and the surgical approach may need to be altered.

T2 tumors are larger (>3 cm) and may involve the main bronchus but should not be closer than 2 cm to the carina. Involvement of the takeoff of the lobar bronchus or main bronchus should be mentioned to allow the surgeon to consider the need for a pneumonectomy or sleeve resection (11). T2 tumors can also have associated atelectasis or pneumonitis that does not include the entire lobe, and these findings should be included in the description of the primary tumor.

T3 tumors can be any size, but they invade the chest wall, diaphragm, mediastinal pleura, or parietal pericardium, or are within 2 cm of the carina, or have associated atelectasis or pneumonitis of the entire lobe. While these lesions can be technically challenging for surgeons to resect, none of the previously mentioned criteria preclude surgical resection (Fig 1). It should be reported if the tumor is less than 2 cm from the carina because this might require a carinal pneumonectomy, which is a formidable procedure with best results at centers where physicians, support personnel, and facilities have experience managing patients undergoing this complicated procedure (12). The extent of chest wall invasion is also important information to convey to the surgeon because more extensive surgical techniques are needed for larger tumors. For this reason, the radiologist should make every attempt to evaluate for invasion of these structures. If the tumor extends into the soft tissues of the chest wall, if rib destruction has occurred, or if the tumor encases mediastinal structures, such invasion is often obvious at imaging. In other cases, the findings on radiographs, CT scans, and MR images are suggestive, but not definitive, in determining invasion of these structures (13,14).

View larger version (76K): [in this window] [in a new window] [Download PPT slide]   Figure 1. Transverse CT image in a 62-year-old woman with a central tumor in the left upper lobe and atelectasis (T3 tumor) shows tumor (T) occluding (arrow) the left upper lobe bronchus and surrounding the pulmonary artery. At surgery, the tumor was invading the pericardium and required intrapericardial pneumonectomy. At 1-year after surgery, there was no local recurrence, but the patient developed distant metastasis.

Invasion of the primary tumor into the mediastinum is also important to assess in patients considered for surgical resection. Sensitivity, specificity, and accuracy of CT and MR imaging for confirming invasion into the mediastinum have been reported to be 40%–84%, 57%–94%, and 56%–89% for CT and 59%–90%, 75%–87%, and 50%–93% for MR imaging, respectively (18–22). But findings reported to indicate mediastinal invasion (eg, contact between the tumor and the mediastinum extending more than 3 cm, obliteration of the fat plane between the mediastinum and the tumor, or the tumor contacting more than 90° of the aortic circumference) are unreliable in differentiating invasion from anatomic contiguity (20,23,24). Although the use of PET and PET/CT with ¹⁸F FDG has been shown to improve staging of lung cancer (4,25), an improvement in determining the extent of chest wall invasion by the primary tumor has not been reported.

T4 tumors involve the mediastinum, heart, great vessels, trachea, esophagus, vertebral body, or carina, or have a malignant pleural or pericardial effusion. Local control of disease is one of the most important aspects of care to improve survival, and, therefore, surgical resection of some T4 tumors may be undertaken (Fig 2). As long as the patients are in good physical condition and do not have evidence of systemic disease, some of these cases represent the one subset of patients with clinical stage IIIB who remain good candidates for surgery (26).

View larger version (138K): [in this window] [in a new window] [Download PPT slide]   Figure 2. Transverse CT image in a 51-year-old woman with squamous cell carcinoma of the left lower lobe shows large left lower lobe mass (M) invading the left atrium (arrows). Patient underwent left pneumonectomy and left atrial resection and reconstruction. Patient was alive at 1 year, with evidence of small contralateral lung metastasis.

It is important to note that in the most recent version of the International Staging System, primary tumors associated with satellite nodules in the same lobe are also classified as T4 (3). However, classification of these lesions as T4 may imply a worse prognosis than is warranted, and some authors (33,34) advocate that patients with satellite nodules undergo definitive resection if no other contraindications to surgery exist. Of note, nodules in another lobe are considered metastatic disease (M1), although many of these are resected because it is not possible to distinguish metastases from synchronous primary tumors, which can be cured with surgery alone (35).

Establishing the presence of a malignant pleural effusion can be difficult, because results of cytologic evaluation are positive in only approximately 66% of patients who have a malignant pleural effusion at presentation (36). In the absence of cytologically positive fluid, a clinical T4 staging classification is still assigned if the clinical suspicion is that the effusion is from the underlying cancer (3,37), but emphasis can be made that diagnostic confirmation is still required. Results of some studies (38–41) have suggested that ¹⁸F FDG PET can aid in characterizing pleural disease. In the report of a study of 92 patients with NSCLC and pleural abnormalities at CT, Schaffler et al (41) reported that sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of FDG PET were 100%, 71%, 63%, 100%, 80%, respectively, and when combined with CT these results were improved to 100%, 76%, 67%, 100%, and 84%, respectively. Therefore, it is important for the radiologist to indicate the presence of pleural disease in the radiologic report and to correlate CT and PET images when available.

One group of T4 tumors that are amenable to resection includes superior sulcus tumors (also known as Pancoast tumors). Findings that prevent surgical resection of a Pancoast tumor include involvement of the brachial plexus and extension of the tumor into the spinal canal. A multidisciplinary treatment approach to this type of tumor can result in improved survival (42). However, tumor invasion into the vertebral body is no longer a contraindication to resection (31,32,43,44). In the past, patients with Pancoast tumors involving the vertebral body were reported to have a 5-year survival rate of less than 10% (45). Gandhi et al (31) showed good intermediate-term survival (2-year survival, 54%) in patients with vertebral body invasion who underwent vertebral resection with reconstruction; if complete resection with negative margins was achieved, the 2-year survival improved to 80%. In another study (46), en bloc resection of tumors involving the intervertebral foramina or vertebral body in which resection was complete led to a 3-year survival rate of 39% and 5-year survival rate of 20%. For this reason, the surgeon needs to know the level of vertebral body involvement, because more than 50% involvement means a poor prognosis, and thus these cases are considered inoperable (31). MR imaging is most useful in patients with superior sulcal tumors to determine whether the tumor has invaded the brachial plexus, subclavian vessels, or vertebral bodies (18,47).

Nodal Disease
After evaluation of the primary tumor, radiologists need to determine the presence and location of regional lymph node metastases (N descriptor). Accurate assessment of lymph nodes of the mediastinum is essential in selecting the best treatment and prognosis in patients with NSCLC (48). To ensure a consistent and standardized description of nodal metastases, the American Thoracic Society has defined nodal stations in relation to anatomic structures or boundaries that can be identified before and during thoracotomy. Although the American Thoracic Society description of nodal stations is the most commonly used system, others, such as that proposed by Naruke and colleagues (49), are also in use.

The presence of nodes within the ipsilateral peribronchial region or hilum indicates N1 disease. N1 lymph nodes can usually be resected at surgery, but technically they are more difficult to remove if they involve the pulmonary artery. Involvement of the pulmonary artery may require pneumonectomy rather than lobectomy to obtain clear surgical margins.

The presence of mediastinal adenopathy critically affects the resectability of the disease. Ipsilateral mediastinal or subcarinal adenopathy constitutes N2 disease and may be resectable. However, contralateral mediastinal adenopathy or any scalene or supraclavicular adenopathy constitutes N3 disease, and patients with such disease are not candidates for surgery. Unfortunately, CT is not ideal for evaluating the mediastinum for lymph node metastasis. Two metaanalyses of staging NSCLC at CT indicate the limitations of using CT for staging the mediastinum (50,51). The earlier review by Dales et al (50) of 42 studies reported a combined sensitivity of 83%, specificity of 82%, and accuracy of 80%. Dwamena et al (51) later reviewed 29 studies and reported a combined sensitivity of 60%, specificity of 77%, and accuracy of 75%. Although the use of multi–detector row CT has improved imaging of the chest, an improvement in chest CT for staging of lung cancer is not expected; this is largely because CT scans only show the size, shape, and location of mediastinal lymph nodes.

Lymph node size at chest CT is the criterion most often used for distinguishing normal from abnormal nodes. A short-axis nodal diameter of 10 mm is considered the upper limit of normal nodal size (52,53). More accurate evaluation of nodal metastases has been reported when different size criteria are used for specific mediastinal regions of the American Thoracic Society system (54). For example, the use of a short-axis nodal diameter of 13 mm as the upper limit for normal nodes in the subcarinal, precarinal, and tracheobronchial regions and of 10 mm for the remaining regions can reduce the number of false-positive results (54). With this method, sensitivity, specificity, and accuracy for detection of N2 disease were 69%, 94%, and 86%, respectively, compared with 74%, 77%, and 86%, respectively, when the standard nodal diameter of 10 mm was used as the upper limit for normal nodes (54).

The evaluation of mediastinal lymph nodes may also be improved by comparing the largest nodes in the expected lymphatic drainage of the tumor with nodes in the rest of the mediastinum (55). By using a short-axis nodal diameter of more than 10 mm and a difference of more than 5 mm between this node and the largest node in the rest of the mediastinum to indicate an abnormal node, the number of false-positive results can be decreased (55). The specificity (99%) and positive predictive value (95%) that resulted from using this method are higher than the specificity (90%) and positive predictive value (73%) that resulted from using an absolute nodal size of more than 10 mm or the specificity (95%) and positive predictive value (77%) that resulted from using different abnormal nodal sizes depending on the regions with American Thoracic Society system (55).

PET imaging with FDG has been shown to be more accurate (reported accuracy, 81%–96%) than CT and MR imaging in the detection of nodal disease (51,56–61). PET is also useful in the differentiation of hyperplastic nodes from metastatic nodes and in the detection of metastasis within normal-sized nodes (59,60,62). In a metaanalysis of nodal staging, the sensitivity was 79% and specificity was 91% for PET compared with 60% and 77%, respectively, for CT (51). One of the most exciting advances in FDG PET imaging is fused PET/CT, which allows better anatomic resolution of FDG-avid lesions (Fig 3 ). PET/CT has recently been reported to be more accurate than PET or CT alone for staging in patients with NSCLC (4,25). In a study of NSCLC in 29 patients performed by Antoch et al (25), the accuracy for depicting metastatic mediastinal lymph nodes was 63% for CT alone, 89% for PET alone, and 93% for PET/CT, while the sensitivity and specificity were 89% and 94% for PET/CT, 89% and 89% for PET, and 70% and 59% for CT, respectively (25). For the evaluation of mediastinal nodes, there were statistically significant differences between results for CT and PET and between results for CT and PET/CT but not between results for PET and PET/CT (25).

View larger version (145K): [in this window] [in a new window] [Download PPT slide]   Figure 3a. Images in a 58-year-old man with NSCLC of the right upper lobe. (a) Transverse CT image of the lower neck shows a small lymph node (arrow) along the left internal jugular vein that could easily be overlooked. (b) Fused PET/CT image shows increased activity of the lymph node (arrow), indicative of metastatic disease.

View larger version (86K): [in this window] [in a new window] [Download PPT slide]   Figure 3b. Images in a 58-year-old man with NSCLC of the right upper lobe. (a) Transverse CT image of the lower neck shows a small lymph node (arrow) along the left internal jugular vein that could easily be overlooked. (b) Fused PET/CT image shows increased activity of the lymph node (arrow), indicative of metastatic disease.

The value of CT in the assessment of lymph nodes of the mediastinum in patients with T1 lesions is also controversial among thoracic surgeons (64). Patients with a T1 lesion on chest radiographs would most likely undergo CT because of its ability to help characterize some lesions as benign and to enable assessment for other pulmonary lesions. If the CT scan shows no mediastinal disease, many surgeons do not routinely perform mediastinoscopy for peripheral T1 lesions, because the prevalence of nodal metastasis associated with T1 tumors is reported to be 5%–15% (7), and because surgeons view the procedure as overly invasive, with substantial morbidity and even mortality (64). However, other surgeons advocate mediastinoscopy in all patients with T1 lesions (64–66) regardless of CT findings. Seely et al (65) reported a prevalence of up to 21% for nodal metastasis in 104 patients with T1 tumors, although CT was reported to have a sensitivity of 77% in ruling out such metastasis. Another group (59) has advocated that mediastinoscopy be avoided in patients with potentially resectable lung cancer and PET images that show normal uptake of FDG in mediastinal nodes. Even with the improved staging that is possible by using PET, controversy regarding its usefulness exists because the true cost-effectiveness of PET and PET/CT are unknown (67). Despite this controversy, radiologists should continue to critically evaluate the mediastinum with available imaging techniques and work to improve the ability of these techniques to enable accurate staging of the mediastinum. In addition, emphasis in the report that pathologic confirmation is needed may be helpful to the treating clinician.

Metastatic Disease
Metastases to the adrenal glands, liver, brain, bones, and lymph nodes may be present in patients with NSCLC at diagnosis. However, the decision to perform extrathoracic imaging to detect metastatic disease is not uniform. Results of previous studies have shown that patients with early stage (T1–2, N0) NSCLC have a less than 1% incidence of occult metastasis; therefore, routine evaluation of these patients for metastasis is not warranted (68,69). In addition, metastatic disease is detected three times as often in patients with clinical and laboratory evidence of distant disease as it is in those without such disease. Nevertheless, data indicate that asymptomatic patients continue to undergo radiologic examinations for metastatic evaluation (70,71). Other clinicians will conduct early evaluation for metastasis if the histologic analysis of the tumor reveals adenocarcinoma, because this type of tumor more commonly manifests metastasis (72); in contrast, squamous cell carcinoma metastases tend to occur later in the disease. Therefore, it is helpful if radiologists know the preference of referring physicians in regard to evaluation for metastatic disease and any clinical information about the patient, because such information can help with image interpretations.

The adrenal glands should be evaluated during diagnostic chest CT because adrenal metastases are common, with a reported detection rate as high as 20% at initial presentation (73–76). Because adrenal adenomas are also common, occurring in 2%–10% of the general population (6), they should be distinguished from metastases. If an adrenal mass has an attenuation value of less than 10 HU on an unenhanced CT scan or there is a reduction of 50% or more Hounsfield units on a 10-minute delayed CT scan, then it is considered benign (77,78). For those adrenal lesions that remain indeterminate at CT, MR imaging or PET is useful for further characterization. Chemical shift analysis with MR imaging can help determine if an adrenal lesion is benign. Evaluation at CT and MR imaging is based on there being adequate fat content in the adrenal mass. FDG PET is also an effective method of evaluating adrenal masses because it demonstrates metabolic activity rather than fat content of the mass. If these methods do not demonstrate whether the mass is benign or malignant, biopsy of the adrenal lesion is warranted. Identification of adrenal metastasis is particularly important because improved survival has been associated with resection of isolated adrenal metastases regardless of whether they are synchronous or metachronous lesions (79). While surgical resection is most common, radiofrequency ablation of adrenal metastasis is a new technique that is performed by some radiologists and shows early promise (80). Therefore, radiologists should carefully report the presence and extent of adrenal metastasis.

The necessity of evaluating the liver at initial presentation is controversial. Of course, if the adrenal glands are included in diagnostic CT scanning, then a substantial portion of the liver will also be imaged. Therefore, any suspicious lesion within the liver warrants further evaluation with dedicated liver CT, MR imaging, or biopsy.

Although metastatic disease of the brain has been reported to be present in up to 18% of patients with NSCLC (72,81–83), the large majority will have signs and symptoms of a neurologic abnormality, and, therefore, routine CT of the brain is not recommended (69,84). As with adrenal metastases, an improved survival is associated with resection of solitary brain metastasis in patients with NSCLC (85,86). The best results occur with metachronous lesions, whereas resection of NSCLC lesions and synchronous brain metastasis is controversial (85).

Patients with skeletal metastases are usually symptomatic or have laboratory test result abnormalities suggestive of bone metastases (72). Because occult skeletal metastases are only occasionally detected at imaging, it is recommended that bone radiography, technetium 99m methylene diphosphonate bone scintigraphy, and MR imaging be performed only to evaluate a history of focal bone pain or an elevated alkaline phosphatase level (69,72,87–89). This recommendation may soon be challenged, however, because results of a previous study (90) of FDG PET in NSCLC showed that 13% of patients, many of whom were asymptomatic, had skeletal metastases.

The controversy surrounding extrathoracic staging in patients with NSCLC may become moot, because whole-body imaging with FDG PET is being routinely performed for staging of NSCLC in patients. Even though the study is performed for a thoracic malignancy, the study commonly involves the whole body. Therefore, regardless of clinical or laboratory evidence, additional unsolicited information about the patient is provided. Results of a number of nonrandomized studies have revealed potential benefits of FDG PET and PET/CT (70). Results of several studies have already shown that FDG PET staging of intra- and extrathoracic disease with a single examination enables detection of occult extrathoracic metastases in 11%–14% of patients initially selected for curative resection and enables alteration of treatment in as many as 40% of patients (51,57,91). Unfortunately, the number of randomized controlled studies performed to evaluate FDG PET for staging in patients with NSCLC is limited (92,93). In a randomized study by van Tinteren et al (92), the addition of PET imaging to conventional work-up in patients suspected of having NSCLC led to a reduction in futile thoracotomies, from 41% to 21%. In that study, futile surgery occurred in both groups regardless of clinical stage, but it occurred much less when PET was included: conventional work-up (46%) versus conventional work-up and PET (25%) for stage I–II disease, and conventional work-up (29%) versus conventional work-up and PET (11%) for stage III disease.

Perhaps more important, despite the additional cost of adding PET to the conventional work-up, Verboom et al (94) showed in a cost analysis of the trial by van Tinteren et al that there was an overall lower total cost of care for the group undergoing conventional work-up and PET compared with a group undergoing conventional work-up alone. A more recent randomized study by Viney et al (93) of PET in stage I and II NSCLC showed a change in stage in 20% of patients, but there was no reduction in the number of thoracotomies because the change was not enough to preclude surgery. However, at the time of this study the treatment for N2 disease was surgery, whereas the authors stated that if neoadjuvant chemotherapy or chemoradiotherapy been used, up to 20% of patients would have been treated differently (93). Authors of both randomized studies acknowledge that because of potential false-positive results from PET, more invasive procedures, such as mediastinoscopy, may be needed to confirm suspected disease.

Another tool in the diagnosis and staging of lung cancer that has been used by some surgeons is video-assisted thorascopic surgery (95,96). This procedure can be a safe and effective method for diagnosis of lung cancer, biopsy of suspected pulmonary metastasis, evaluation of the pleura for metastasis, and staging of mediastinal lymph nodes (97,98). Some surgeons even perform lobectomy by means of video-assisted thorascopic surgery for stage I lung cancer because of reported shorter recovery times and less postoperative pain (99,100). However, these advantages have been reported to be lost within 2 weeks after surgery when compared with muscle sparing or limited thoracoscopic procedures (101). Therefore, the accurate description of tumor location relative to the peripheral surface, possible pleural involvement, and mediastinal lymph nodes is important to guide surgeons who are performing video-assisted thorascopic surgery.

Postoperative Imaging
Radiologic evaluation in patients continues after surgical resection of NSCLC. In the immediate postoperative period, portable chest radiography is performed to assess for lobar collapse, tension pneumothorax, pulmonary edema, or other acute processes that can be catastrophic if left untreated. Any acute findings in these patients should be discussed immediately with the surgeon. Stabilization of patients, both clinically and radiographically, will allow them to be discharged home. For patients who have undergone pneumonectomy, most surgeons prefer to have the fluid within the pleural space cover the bronchial stump prior to discharge, so radiologists should note this event when it occurs.

In the weeks and months after surgery, any postoperative pneumothorax will decrease in size and will usually be replaced by fluid or compensatory expansion of the remaining lung. Patients with pneumonectomy will usually have complete opacification of the hemithorax with shift of the mediastinum into the surgical side within 6–8 months (102). Usually, variable amounts of fluid and air remain within the pneumonectomy space, and rarely will there be complete absorption of fluid (103). Occasionally, a gas-fluid level (hydropneumothorax) may persist, and, as long as it remains stable or the patient is asymptomatic, it is not clinically important (104). However, an increase in the gas component of the postoperative hydropneumothorax should raise concern for a bronchopleural fistula. On rare occasions, patients with resected tumors that involved the spinal structures can develop a bronchopleural subarachnoid fistula.

Postoperative radiologic follow-up of patients after the immediate postoperative period varies according to the referring surgeon. Follow-up examinations typically include chest radiography and chest CT, with extrathoracic imaging ordered on the basis of clinical symptoms that are suggestive of potential metastatic disease (105). However, results of some studies suggest little value to close short-term follow-up (106), so it is important for radiologist to understand the ordering patterns for radiology examinations by the referring surgeons. Knowing that a surgeon does not usually order CT for follow-up can suggest that a recurrent tumor is suspected when this study is ordered. When evaluating follow-up images, radiologists should be diligent in assessing the surgical site for evidence of recurrent disease and the remaining examination images for metastatic disease.

	RADIATION ONCOLOGIST'S PERSPECTIVE

TOP ABSTRACT INTRODUCTION SURGEON'S PERSPECTIVE RADIATION ONCOLOGIST'S... MEDICAL ONCOLOGIST'S PERSPECTIVE RADIOLOGIST IN MULTIDISCIPLINARY... CONCLUSION References

 
Pretreatment Imaging
More than 60% of patients with lung cancer will undergo radiation therapy at some point in their disease: as initial treatment in 45% of patients and as palliative treatment in 17% (107). According to recent estimates, almost 174 000 patients will have developed lung cancer in the United States in 2004 (2), which means over 100 000 Americans were irradiated for lung cancer last year.

Radiation therapy for a curative intent is used in patients with early stage (stage I–II) NSCLC who are not surgical candidates because of medical comorbidities or patient refusal of surgery (108). Use of radiation therapy alone in this group has a median survival time of 30 months and a 5-year survival rate of 42%. The difference in survival rate between radiation therapy and surgery in patients with early-stage lung cancer is complex but is affected by medical comorbidities, incomplete staging (clinical, not pathologic), and variations in treatment delivery and supportive care. For locally advanced and inoperable lung cancer, chemotherapy is used in conjunction with radiation therapy to improve patient survival (109–111). Preoperative concurrent chemoradiotherapy in locally advanced lung cancer has shown promising results but remains controversial and needs further validation (112). Because the determination of disease extent in patients undergoing radiation therapy is often based on clinical staging, imaging examinations and the radiologist's interpretation are important aspects of treatment planning and in caring for these patients.

Initially, the radiologic evaluation and staging in patients undergoing radiation therapy are similar to the evaluation and staging in patients undergoing surgery. As with surgeons, radiation oncologists need to understand the size and location of the primary tumor, as well as any potential regional lymph node involvement. The proximity of the tumor to critical structures should be stressed, particularly the proximity to structures such as the lung, spinal cord, esophagus, and heart, where radiation tolerance imposes dose-volume constraints. Evaluation of tumor size is also important but can be difficult if there is associated atelectasis or obstructive pneumonitis. However, underestimation of tumor extent can lead to high local recurrence rates, whereas overestimation can lead to destruction of normal tissue and complications. Therefore, technically sound CT examinations should be performed with thin-section evaluation of the primary lesion. Radiologists should assist radiation oncologists by describing tumor margins and the extent of tumor invasion into adjacent structures. If invasion is not definite, this uncertainty should also be noted.

The pretreatment needs of radiation oncologists differ from those of other oncologic specialists in two important respects: First, radiation oncologists must be able to determine the margins of a visible tumor. Second, radiation oncologists need to integrate multidisciplinary imaging (CT and FDG PET, CT and MR imaging, or other types of biologic imaging) to define geometrically accurate targets needed for treatment planning.

The goal of radiation therapy is to adequately treat the tumor while limiting damage to the surrounding tissues. Although current imaging techniques cannot help determine the true microscopic limits of tumors, defining the radiologically visible margins is important. To account for microscopic disease that cannot be imaged, the International Commission of Radiation Units and Measurements has defined several volumes that are important for the modern treatment of lung cancer (113,114): the gross tumor volume, which is the volume of tumor that is visible with any imaging modality; the clinical target volume, which is the volume that is likely to contain microscopic disease on the basis of reported patterns of recurrence; and the planning target volume, which includes the clinical tumor volume with a margin to account for daily setup error and target motion. Diagnostic radiologists can be very helpful to the radiation oncologist by delineating the gross tumor margin.

Determination of tumor margins is becoming more important in radiation therapy because many centers now use conformal radiation therapy, which is a technique that uses multiple radiation beams to generate dose distributions that conform tightly to target volumes and limit damage to normal structures. As the area of treatment surrounding a tumor becomes smaller, it is critical that delineation of the margins be accurate. Poor determination of tumor margins can result in inadequate treatment or excessive damage to surrounding tissues.

Moreover, determination of tumor margins is not limited to lung lesions. Because most radiation therapy treatment planning CT for lung cancer is performed without contrast material, radiologists who are familiar with mediastinal anatomy can also assist in defining lymph nodes that should be included in treatment. The radiologist's interpretation of potential nodal disease in the mediastinum is very important in these patients, because they will not undergo mediastinoscopy for assessment of metastatic disease. Radiologists should work closely with radiation oncologists to appropriately identify lymph nodes that are suspicious for metastatic disease so that treatment planning for radiation therapy includes all potential malignancy. When interpreting diagnostic chest CT scans, radiologists should indicate the location and size of all adenopathy of the chest, including that of the mediastinum, hili, supraclavicular, paraesophageal, axillary, and retrocrural stations to provide guidance to the radiation oncologist.

The rapid evolution of fused PET/CT imaging is proving to be a valuable tool in treatment planning for patients with NSCLC who are to undergo radiation therapy. In one study (115), use of PET/CT resulted in alterations in the treatment plan in more than 50% of patients with NSCLC when compared with CT alone. Fused imaging with use of FDG allows the differentiation of suspected metastatic disease from benign lesions, such as brown fat, and also helps the radiologist to differentiate recurrent tumor from radiation fibrosis (Fig 4). While fused imaging has been rapidly integrated into radiation therapy treatment planning, training of radiation oncologists and physicists in the use of this modality has not kept pace. Radiologists familiar with this emerging technology should be available to consult with radiation oncologist. For example, although there is improved sensitivity of FDG PET, tumor edges cannot be accurately identified with PET. The range of positrons in the lung, the large voxel size, the use of data filters (often unsuspected by the radiation oncologists), and tumor misregistration in fused imaging all contribute to poor edge depiction. There is also a need for radiology physicists to train radiation therapy physicists. As fused PET/CT is used more often for treatment planning, the physicists involved in radiation treatment planning have to become more familiar with fused PET/CT imaging.

View larger version (94K): [in this window] [in a new window] [Download PPT slide]   Figure 4a. Images in a 64-year-old woman with adenocarcinoma of the left upper lobe treated with chemoradiation. Follow-up CT results could not exclude persistent tumor. (a) Transverse CT image shows a masslike appearance (arrows) to the radiation fibrosis in the left upper lobe that was concerning for recurrent tumor. The air bronchograms typically seen in radiation fibrosis are not evident. (b) Coronal CT reconstruction shows the extent of the mass (arrow) above the aortic arch (A). (c) Coronal fused PET/CT image obtained to evaluate the mass shows no increased FDG uptake (arrow), indicating no malignancy. Follow-up imaging for 9 months continued to show no increased FDG uptake within the mass, thus confirming stable radiation fibrosis. A = aortic arch.

View larger version (87K): [in this window] [in a new window] [Download PPT slide]   Figure 4b. Images in a 64-year-old woman with adenocarcinoma of the left upper lobe treated with chemoradiation. Follow-up CT results could not exclude persistent tumor. (a) Transverse CT image shows a masslike appearance (arrows) to the radiation fibrosis in the left upper lobe that was concerning for recurrent tumor. The air bronchograms typically seen in radiation fibrosis are not evident. (b) Coronal CT reconstruction shows the extent of the mass (arrow) above the aortic arch (A). (c) Coronal fused PET/CT image obtained to evaluate the mass shows no increased FDG uptake (arrow), indicating no malignancy. Follow-up imaging for 9 months continued to show no increased FDG uptake within the mass, thus confirming stable radiation fibrosis. A = aortic arch.

View larger version (100K): [in this window] [in a new window] [Download PPT slide]   Figure 4c. Images in a 64-year-old woman with adenocarcinoma of the left upper lobe treated with chemoradiation. Follow-up CT results could not exclude persistent tumor. (a) Transverse CT image shows a masslike appearance (arrows) to the radiation fibrosis in the left upper lobe that was concerning for recurrent tumor. The air bronchograms typically seen in radiation fibrosis are not evident. (b) Coronal CT reconstruction shows the extent of the mass (arrow) above the aortic arch (A). (c) Coronal fused PET/CT image obtained to evaluate the mass shows no increased FDG uptake (arrow), indicating no malignancy. Follow-up imaging for 9 months continued to show no increased FDG uptake within the mass, thus confirming stable radiation fibrosis. A = aortic arch.

Imaging performed within 3 months of treatment will often reveal ground-glass opacities, which indicates radiation pneumonitis. On occasion, pneumonitis may appear nodular and simulate metastatic nodules (Fig 5), but radiologists need to appropriately categorize this as pneumonitis. Typically, the nodules of radiation pneumonitis will be within the lung that underwent radiation therapy and are irregular or poorly marginated (116,117). At follow-up imaging, the nodules will be seen to coalesce into areas of consolidation and will eventually become a component of the radiation fibrosis. Radiation fibrosis consists of a well-defined area of consolidation associated with volume loss and bronchiectasis. Commonly, the development of fibrosis will progress slowly over the 3–12 months after radiation therapy ends and will stabilize within 2 years. The majority of fibrosis will occur within the first 12 months and will have a well-defined border (116,117). Radiation oncologists will usually need to confirm stability of the radiation fibrosis with imaging and anticipate radiologists to report these findings.

View larger version (77K): [in this window] [in a new window] [Download PPT slide]   Figure 5a. Transverse CT images in a 36-year-old man with a right lower lobe moderately differentiated squamous cell carcinoma. The patient underwent chemo- and radiation therapy. (a) Mediastinal (window: width, 599 HU; level, 55 HU) image shows a right lower lobe mass (M) occluding the bronchus intermedius. (b) Lung (window: width, 1500; level, –600) image at baseline shows minimal surrounding pulmonary opacities (arrows). M = mass. (c) Image obtained at 8-month follow-up shows scattered nodules (arrowheads) and poorly defined opacities in the right lower lobe. Note the sharp demarcation (arrows) of the irradiated lung from normal lung and that the nodules and opacities are within the demarcated area of irradiation. These nodules should not be confused for metastatic nodules because they are within the radiation field. (d) Image obtained at 14 months shows well-organized radiation fibrosis with sharp demarcation (arrows) between fibrosis and normal lung. Nodules are no longer evident and there are no metastatic nodules.

View larger version (107K): [in this window] [in a new window] [Download PPT slide]   Figure 5b. Transverse CT images in a 36-year-old man with a right lower lobe moderately differentiated squamous cell carcinoma. The patient underwent chemo- and radiation therapy. (a) Mediastinal (window: width, 599 HU; level, 55 HU) image shows a right lower lobe mass (M) occluding the bronchus intermedius. (b) Lung (window: width, 1500; level, –600) image at baseline shows minimal surrounding pulmonary opacities (arrows). M = mass. (c) Image obtained at 8-month follow-up shows scattered nodules (arrowheads) and poorly defined opacities in the right lower lobe. Note the sharp demarcation (arrows) of the irradiated lung from normal lung and that the nodules and opacities are within the demarcated area of irradiation. These nodules should not be confused for metastatic nodules because they are within the radiation field. (d) Image obtained at 14 months shows well-organized radiation fibrosis with sharp demarcation (arrows) between fibrosis and normal lung. Nodules are no longer evident and there are no metastatic nodules.

View larger version (122K): [in this window] [in a new window] [Download PPT slide]   Figure 5c. Transverse CT images in a 36-year-old man with a right lower lobe moderately differentiated squamous cell carcinoma. The patient underwent chemo- and radiation therapy. (a) Mediastinal (window: width, 599 HU; level, 55 HU) image shows a right lower lobe mass (M) occluding the bronchus intermedius. (b) Lung (window: width, 1500; level, –600) image at baseline shows minimal surrounding pulmonary opacities (arrows). M = mass. (c) Image obtained at 8-month follow-up shows scattered nodules (arrowheads) and poorly defined opacities in the right lower lobe. Note the sharp demarcation (arrows) of the irradiated lung from normal lung and that the nodules and opacities are within the demarcated area of irradiation. These nodules should not be confused for metastatic nodules because they are within the radiation field. (d) Image obtained at 14 months shows well-organized radiation fibrosis with sharp demarcation (arrows) between fibrosis and normal lung. Nodules are no longer evident and there are no metastatic nodules.

View larger version (110K): [in this window] [in a new window] [Download PPT slide]   Figure 5d. Transverse CT images in a 36-year-old man with a right lower lobe moderately differentiated squamous cell carcinoma. The patient underwent chemo- and radiation therapy. (a) Mediastinal (window: width, 599 HU; level, 55 HU) image shows a right lower lobe mass (M) occluding the bronchus intermedius. (b) Lung (window: width, 1500; level, –600) image at baseline shows minimal surrounding pulmonary opacities (arrows). M = mass. (c) Image obtained at 8-month follow-up shows scattered nodules (arrowheads) and poorly defined opacities in the right lower lobe. Note the sharp demarcation (arrows) of the irradiated lung from normal lung and that the nodules and opacities are within the demarcated area of irradiation. These nodules should not be confused for metastatic nodules because they are within the radiation field. (d) Image obtained at 14 months shows well-organized radiation fibrosis with sharp demarcation (arrows) between fibrosis and normal lung. Nodules are no longer evident and there are no metastatic nodules.

View larger version (120K): [in this window] [in a new window] [Download PPT slide]   Figure 6a. Transverse CT images in a 72-year-old woman with adenocarcinoma of the right upper lobe who underwent radiation therapy. (a) Image obtained 11 months after completion of therapy shows radiation fibrosis (white arrows). Note patency of the air bronchograms (black arrows). (b) Image obtained 1 month later shows filling in of the air bronchograms with soft-tissue opacity (arrows) indicating recurrent tumor.

View larger version (116K): [in this window] [in a new window] [Download PPT slide]   Figure 6b. Transverse CT images in a 72-year-old woman with adenocarcinoma of the right upper lobe who underwent radiation therapy. (a) Image obtained 11 months after completion of therapy shows radiation fibrosis (white arrows). Note patency of the air bronchograms (black arrows). (b) Image obtained 1 month later shows filling in of the air bronchograms with soft-tissue opacity (arrows) indicating recurrent tumor.

View larger version (175K): [in this window] [in a new window] [Download PPT slide]   Figure 7a. Transverse CT images in a 53-year-old woman with adenocarcinoma of the right upper lobe who was not a candidate for surgery because of comorbid disease. Three-dimensional conformal radiation therapy was performed. (a) Baseline image shows a lobulated lesion (arrows) of the right upper lobe. (b) Image obtained 37 months after completion of radiation therapy shows an atypical pattern of radiation fibrosis of a small nodule (arrow) associated with a linear scar. Without prior knowledge of treatment, this could be misinterpreted as a pulmonary nodule arising from a scar.

View larger version (162K): [in this window] [in a new window] [Download PPT slide]   Figure 7b. Transverse CT images in a 53-year-old woman with adenocarcinoma of the right upper lobe who was not a candidate for surgery because of comorbid disease. Three-dimensional conformal radiation therapy was performed. (a) Baseline image shows a lobulated lesion (arrows) of the right upper lobe. (b) Image obtained 37 months after completion of radiation therapy shows an atypical pattern of radiation fibrosis of a small nodule (arrow) associated with a linear scar. Without prior knowledge of treatment, this could be misinterpreted as a pulmonary nodule arising from a scar.

The role of PET in the follow-up of patients treated with radiation therapy is evolving (132). PET imaging following radiation therapy has been shown to be useful for monitoring therapy. In a study by Ryu et al (133) on evaluating neoadjuvant chemoradiotherapy in NSCLC, PET had high sensitivity but limited specificity in the detection of residual tumor in the primary tumor and had high specificity but limited sensitivity in the restaging of mediastinal lymph nodes. Radiologist also need to be aware that radiation pneumonitis can also show increased FDG activity on PET images.

	MEDICAL ONCOLOGIST'S PERSPECTIVE

TOP ABSTRACT INTRODUCTION SURGEON'S PERSPECTIVE RADIATION ONCOLOGIST'S... MEDICAL ONCOLOGIST'S PERSPECTIVE RADIOLOGIST IN MULTIDISCIPLINARY... CONCLUSION References

 
In general, chemotherapy is performed in patients with stage III and IV NSCLC who are not surgical candidates, although subsets of patients undergo chemotherapy prior to surgery and in conjunction with radiation therapy to improve survival. The radiologic evaluation in patients who are treated by medical oncologists is similar to that in patients who are treated by surgeons and radiation oncologists, but the information provided to medical oncologists is different in important ways. Patients treated by medical oncologists generally have recurrent or metastatic disease, and the treatment they receive is often not for local control of the primary tumor but for systemic disease. For this reason, determination of tumor location and margins is only important if not having this information puts the patient at potential risk for a complication or potentially life-threatening event, such as invasion into a vascular structure that may lead to exsanguination. The most important radiologic information that medical oncologists require is the determination of the effectiveness of treatment. In essence, they need to know if the disease is stable, improved, or progressed to determine if the treatment regimen should be continued or changed.

To standardize the methods of determining effectiveness of treatment, uniform criteria for reporting response, recurrence, disease-free interval, and toxicity were adopted at a meeting in 1979 on the Standardization of Reporting Results of Cancer Treatment (134,135). These criteria, known as the World Health Organization criteria, are based largely on tumor measurements in two dimensions (bidimensional), which are obtained by multiplying the longest diameter of the tumor by the greatest perpendicular diameter in the transverse plane. This product is the tumor size, and if there are multiple evaluable tumors, then the sum of the products is obtained to determine the total tumor size. Treatment response is defined as complete response (no evidence of tumor), partial response (decrease in tumor size by at least 50%), stable disease (no change in tumor size), or progressive disease (increase in tumor size by at least 25%) (135).

In 1994, the World Health Organization criteria were revised and guidelines known as the Response Evaluation Criteria in Solid Tumors (RECIST) were proposed (136), whereby tumor measurements are obtained with a single measurement (unidimensional), which is obtained by measuring the longest diameter of the tumor in the transverse plane. This measurement is the tumor size, and if there are multiple tumors used, then the sum of the diameters is used to calculate total tumor size. With RECIST criteria, treatment response is defined as complete response (no evidence of tumor), partial response (decrease in tumor size by at least 30%), stable disease (no change in tumor size), and progressive disease (increase in tumor size by at least 20%). RECIST is now the preferred method of assessing response (136). Regardless of which criteria are used, new metastatic disease is also indicative of progressive disease and needs to be emphasized.

RECIST criteria require the identification of target lesions to be followed for a response to treatment. Target lesions are identified as measurable disease, which by RECIST criteria requires the lesion to be at least twice the size of the section thickness. For instance, if a 5-mm section thickness is used for CT, then a 1-cm lesion is the smallest lesion that can be used to assess tumor response. As many as five lesions per organ can be used, and the longest dimension of each tumor is recorded. This assessment is purely an assessment of tumor size, not function (Fig 8). Bone lesions, although important to identify, are not used for response assessment in chemotherapy trials because changes can be difficult to appreciate in bones. Likewise, nonmeasurable disease, such as a pleural effusion, is noted but not measured. Therefore, it is important for the radiologist to be aware of these criteria and to identify the primary tumor, if present, and any metastatic disease that can serve as measurable disease. For radiologists, it may seem to be a simple matter of determining a change in the size of a particular lesion, but assessment of changes in tumor size can be inaccurate because of technical variations in the imaging method and measurement errors.

View larger version (132K): [in this window] [in a new window] [Download PPT slide]   Figure 8a. Transverse CT images in a 63-year-old man who underwent left lower lobectomy for adenocarcinoma of the left lower lobe. (a) Image obtained 4 months after surgery reveals a right lower lobe lesion confirmed as metastatic disease at biopsy. Baseline measurement is 16.7 mm. (b) Image obtained after completion of first course of chemotherapy shows interval growth. The lesion measures 25.4 mm. According to RECIST criteria, this is 52% growth and compatible with progressive disease.

View larger version (129K): [in this window] [in a new window] [Download PPT slide]   Figure 8b. Transverse CT images in a 63-year-old man who underwent left lower lobectomy for adenocarcinoma of the left lower lobe. (a) Image obtained 4 months after surgery reveals a right lower lobe lesion confirmed as metastatic disease at biopsy. Baseline measurement is 16.7 mm. (b) Image obtained after completion of first course of chemotherapy shows interval growth. The lesion measures 25.4 mm. According to RECIST criteria, this is 52% growth and compatible with progressive disease.

View larger version (131K): [in this window] [in a new window] [Download PPT slide]   Figure 9a. Transverse CT images in a 66-year-old man previously treated for adenocarcinoma of the left upper lobe with new metastatic disease in the right upper lobe. The right upper lobe tumor can be used as a RECIST target lesion to assess response to treatment. (a) Note the lobulated contour (arrow) of the tumor. (b) If one radiologist does not include the small lobule in the measurement, the largest dimension of the tumor is 18.4 mm. (c) If another radiologist measures the same tumor but includes the small projection, the largest dimension is 24.4 mm. This increase in size could be interpreted as progressive disease (33% increase), even though there is no difference. Therefore, a lesion that is not changed on follow-up images may be stable, but interobserver errors in measuring could indicate a change that would result in alteration of treatment.

View larger version (128K): [in this window] [in a new window] [Download PPT slide]   Figure 9b. Transverse CT images in a 66-year-old man previously treated for adenocarcinoma of the left upper lobe with new metastatic disease in the right upper lobe. The right upper lobe tumor can be used as a RECIST target lesion to assess response to treatment. (a) Note the lobulated contour (arrow) of the tumor. (b) If one radiologist does not include the small lobule in the measurement, the largest dimension of the tumor is 18.4 mm. (c) If another radiologist measures the same tumor but includes the small projection, the largest dimension is 24.4 mm. This increase in size could be interpreted as progressive disease (33% increase), even though there is no difference. Therefore, a lesion that is not changed on follow-up images may be stable, but interobserver errors in measuring could indicate a change that would result in alteration of treatment.

View larger version (129K): [in this window] [in a new window] [Download PPT slide]   Figure 9c. Transverse CT images in a 66-year-old man previously treated for adenocarcinoma of the left upper lobe with new metastatic disease in the right upper lobe. The right upper lobe tumor can be used as a RECIST target lesion to assess response to treatment. (a) Note the lobulated contour (arrow) of the tumor. (b) If one radiologist does not include the small lobule in the measurement, the largest dimension of the tumor is 18.4 mm. (c) If another radiologist measures the same tumor but includes the small projection, the largest dimension is 24.4 mm. This increase in size could be interpreted as progressive disease (33% increase), even though there is no difference. Therefore, a lesion that is not changed on follow-up images may be stable, but interobserver errors in measuring could indicate a change that would result in alteration of treatment.

Assessment of response to therapy with PET has also shown promise in the differentiation of viable tumor from fibrosis (133,138–140). In one study (138) of 15 patients who completed chemotherapy, a reduction in FDG activity of 50% or more in the primary tumor was a better predictor of survival than were World Health Organization criteria. In a study of 57 patients, Weber et al (139) found that a reduction in FDG activity (as assessed by means of standardized uptake values) of 20% after one cycle of chemotherapy resulted in a 1-year survival rate of 44% compared with a survival rate of only 10% if tumors did not respond. The results of these early studies suggest an important role for PET in the assessment and management of disease in patients undergoing chemotherapy.

An important aspect of the radiologist's role in treating the patient with NSCLC is the detection of complications from chemotherapy. The most common complications are infections and drug toxicity. Both can manifest fever, malaise, and cough. Patients with cancer are susceptible to infections because they are immunocompromised from their disease or treatment regimens. The lung is the most common site of serious infections in patients with cancer, and causes of infection include bacterial, fungal, or viral sources, all of which can be detected with imaging.

Drug toxicity also has to be considered in these patients and requires a high index of suspicion because it can mimic infections, radiation pneumonitis, or recurrent tumor (141). Drug toxicity manifests radiologically