| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
EDUCATION EXHIBIT |
Department of Radiology, The University Hospital, Cincinnati, Ohio
Lung cancer remains the most common cause of cancer death in the United States, with 173,700 new cases and 164,440 deaths expected in 2004 (1). Despite aggressive surgery, new chemotherapies, innovative stereotactic radiation therapy, and, most recently, radiofrequency ablation, the overall 5-year survival rate has remained 12%–15% for the past several decades (2). Early reports suggest that screening CT for lung cancer may offer the opportunity to detect and treat lung cancer at an early stage. In feasibility studies, 63%–93% of lung cancers detected at initial screening CT and 74%–94% of lung cancers detected at follow-up screening CT were stage I cancers (3). Given the detection of a greater percentage of stage I lung cancer, it may be time to reevaluate the principles guiding lung cancer surgery. Lobectomy has been considered the only definitive surgery for lung cancer because limited resections were associated with a higher risk of local recurrence (4). Unfortunately, patients who survive one lung cancer face a 2.5% annual risk of developing a second primary lung cancer (5). The more lung that is preserved during the first resection, the more surgical options remain should a metachronous primary tumor develop. Furthermore, if we can identify features of "good prognosis" stage IA lung carcinoma, limited surgery may be a viable option for cure.
Lee et al have reviewed some of the prognostic information that can be derived from imaging data by the radiologist as a participant in the multidisciplinary approach to lung cancer. The authors did so with specific questions in mind. For example, does the percentage of ground-glass attenuation correlate with histologic type and therefore have therapeutic and prognostic implications? Can radiologists better stratify tumors within stage IA to more accurately predict the likelihood of nodal or distant metastases and, thus, survival? What is the clinical significance of pure ground-glass attenuation, which is commonly encountered in nonsmokers in screening CT studies from Asia? What is the relationship of pure ground-glass attenuation to fatal lung cancer?
At histologic analysis, pure ground-glass attenuation represents atypical adenomatous hyperplasia or BAC in situ, not the typical invasive adenocarcinoma detected in smokers. Furthermore, doubling times for pure ground-glass attenuation lesions are reported to be 880 days or more, in contrast to typical adenocarcinomas, whose doubling times average 180 days (6). If we assume exponential growth, a 1-cm ground-glass attenuation nodule would take 5 doubling times (12 years) to enlarge to 3 cm (7). Evaluation of the natural history of pure ground-glass attenuation lesions led Kodama et al (8) to conclude that "some pure ground-glass attenuation [lesions] will never progress to clinical disease and would be included in the category of overdiagnosis bias." Currently, only 10% of all smokers die of lung cancer (9). A focus of pure ground-glass attenuation may not affect survival.
Lee et al have discussed the results of several studies in the growing body of literature that correlates the proportion of ground-glass attenuation in stage IA lung cancers with histologic and prognostic significance. Multiple studies report a progressive decrease in the percentage of ground-glass attenuation as the histologic findings progress through a spectrum from atypical adenomatous hyperplasia to BAC with invasive adenocarcinoma (10–20). Noguchi et al (21) classified peripheral adenocarcinoma as follows: Type A, localized BAC; Type B, localized BAC with foci of alveolar collapse; Type C, localized BAC with active fibroblastic proliferation; Type D, poorly differentiated adenocarcinoma; Type E, tubular adenocarcinoma; and Type F, papillary adenocarcinoma with compressive and destructive growth.
The natural history of a Noguchi type A or B lesion (BAC in situ) is uncertain. Although a 100% 5-year survival rate has been reported in affected patients, given the long doubling time of these lesions, 5-year follow-up may be too short to predict the natural history (21). Findings on serial CT scans suggest that Noguchi type A and B lesions may progress to type C adenocarcinomas with a replacement growth pattern over time (22). It is significant that the progression of Noguchi type C adenocarcinomas to more aggressive malignancies is not well established (10). In the original article, Noguchi et al (21) suggested that types D–F do not evolve in a similar stepwise fashion but rather represent de novo adenocarcinomas. Kakinuma et al (23) reported that, in their experience with pure ground-glass attenuation, several lesions were observed to decrease in size at follow-up. This finding has led investigators to suggest that wedge resection at video-assisted thoracoscopic surgery may be a more appropriate, conservative approach to pure ground-glass attenuation lesions (24).
Whereas the Early Lung Cancer Action Project study (25) screened high-risk smokers, the study performed in Nagano, Japan (26) screened a population of 7,847 smokers and nonsmokers aged 40–74 years. The latter study resulted in a 1.1% detection rate for lung cancer with an equal prevalence and incidence of stage IA disease in smokers and nonsmokers. In a study by Kondo et al (27), of the 57 patients with peripheral adenocarcinomas less than 10 mm in size, 78.9% were women and 77.2% were nonsmokers. Kim et al (28) reviewed 224 cases of adenocarcinoma and reported 132 BACs and 92 other adenocarcinomas. Is the distribution of malignancies reported in screening studies from Asia related to genetic or environmental factors? Can the results of these studies, with a disproportionate representation of BAC, be applied to a population of high-risk smokers in the United States?
Finally, as we begin to evaluate stage IA lesions, it becomes clear that this is a diverse group of lesions in terms of size, histologic features, and geographic distribution. Patz et al (29) reported no difference in survival rates based on the size of the stage IA lung cancers. It has been speculated that this may be due to the fact that lung cancers as small as 1 mm are capable of metastasizing in experimental models (30). In another recent series of clinical T1 N0 M0 lung cancers less than 10 mm in size, four of 17 patients had lymph node metastases, which supports the observation that 10%–20% of stage IA lung cancers will be associated with lymph node metastases (5). Other researchers have proposed subdividing T1 N0 M0 lung cancers according to size and have demonstrated improved 5-year survival rates in patients with tumors less than 1–2 cm in diameter (31–37). These studies have not stratified groups of lesions on the basis of nodule attenuation characteristics. In the setting of peripheral adenocarcinomas, percentage of ground-glass attenuation and lesion size less than 1.5 cm correlate with survival rates (12,14,15,19,20). In one study of 96 patients with T1 N0 M0 adenocarcinomas, no lymph node metastases were present in tumors with ground-glass attenuation accounting for more than 50% of tumor volume (38). On the basis of similar data from several other studies, limited resection of these lesions may be justified (8,11,13,39,40).
In the future, it may not be enough to classify a lung cancer less than 3 cm in size as stage IA; with knowledge of the population screened, the radiologist may be expected to describe the size, mean attenuation, and proportion of ground-glass attenuation to soft-tissue attenuation of the lesion in question. In cases of pure ground-glass attenuation, we can predict a favorable prognosis. Particularly in the face of growing concerns regarding overdiagnosis bias, wedge resection may be curative and yet preserve more lung function than a standard lobectomy (9). Further research is necessary to determine whether a limited preoperative evaluation is cost effective for semisolid neoplasms with a ground-glass attenuation component greater than 50%. Finally, for small solid neoplasms, complete radiologic staging with PET and CT is indicated to direct lymph node sampling or biopsy of distant metastatic sites in the 10%–20% of patients who will ultimately prove to have advanced-stage lung cancer.
 |
References |
|---|
 Top References References |
|---|
Department of Radiology and Center for Imaging Science, Samsung Medical Center, Sungkyunkwan University School of Medicine, Seoul, Korea
Department of Medical Physics, Osaka University Graduate School of Medicine, Osaka, Japan
Cancer Screening Technology Division, Research Center for Cancer Prevention and Screening, National Cancer Center, Chiba, Japan
We thank Drs Meyer and Shipley for their commentary on our article.
As was mentioned in both their commentary and our article, peripheral lung cancers, especially adenocarcinomas, are a heterogeneous group of tumors. On the one hand, some tumors are very slow-growing, with a doubling time of 880 days or more (1). These tumors, appearing as pure ground-glass attenuation lesions at CT and representing BAC, account for about 14% of adenocarcinomas (2) and may lead to overdiagnosis bias at lung cancer screening (3). On the other hand, about 24% of T1 lung cancers (nodules less than 3 cm in diameter) show early metastasis to mediastinal lymph nodes or extrathoracic organs, thus resulting in a poor prognosis (4,5). Most of these tumors appear as solid nodules at CT. In the remaining 60% of lung cancers, the tumors may show an ordinary growth pattern with an average doubling time of 180 days, appearing as semisolid nodules (in which ground-glass attenuation accounts for a certain percentage of tumor volume) or solid nodules at CT. Lung cancer screening may be effective in these cancers.
By being aware of the relationship between prognosis and the attenuation value or the volume percentage of ground-glass attenuation in a nodule at CT, radiologists can stratify stage IA tumors, allowing more accurate prediction of the likelihood of nodal or distant metastases and thus, of survival. Furthermore, radiologists may suggest other treatment options; for example, wedge resection for pure ground-glass attenuation nodules of lung cancer (6,7).
In Japanese (8) and Korean (our unpublished data) studies in which screening was performed in the general population—not like the Early Lung Cancer Action Project (ELCAP) study (9), in which screening was performed only in high-risk smokers—smokers and nonsmokers had an equal prevalence and incidence of stage IA disease. Interestingly, however, cancers in smokers showed rapid doubling time, appeared as solid nodules at CT, and were poorly differentiated at pathologic examination (8).
We do not regard the large numbers of BAC cases in Japanese screening studies as disproportionately high. Cancer screening was performed in both smokers and nonsmokers in Japan, and most cases of BAC were found in nonsmoking women (8). Most likely, even in Western countries, there will be many cancers that appear as pure ground-glass attenuation lesions and represent BAC if screening is performed in larger numbers of younger nonsmokers. In the ELCAP study (9), of a total of 233 nodules found at screening, 28 (12%) were pure ground-glass attenuation nodules (described as "nonsolid" nodules). Of these 28 nodules, five (18%) proved to be lung cancer (9,10). This percentage of pure ground-glass attenuation nodules (12%) is not a small number. Our unpublished data on lung cancer screening in Korea revealed that 254 of 4,037 detected nodules (6.3%) were pure ground-glass attenuation nodules. In Japan and Korea, radiologists and chest physicians have a deep insight into such cases from their experience, so that more proved cases of lung cancer with pure ground-glass attenuation may have been seen. Moreover, we have not found any reports describing genetic differences between lung cancers seen in Asian and Western countries.
Lung cancers that manifest as pure ground-glass attenuation lesions and represent BAC (Noguchi types A and B) account for about 14% of all adenocarcinomas. No patients with these cancers have mediastinal nodal or extrathoracic metastasis at initial presentation (11). However, these cancers may show subsequent growth and progress to Noguchi type C adenocarcinoma with an increased proportion of solid component at long-term follow-up without any treatment (12,13). Therefore, wedge resection should be performed for these cancers before they progress to more aggressive Noguchi type C cancers.
|
References |
|---|
|
TopReferencesReferences |
|---|
Related Article
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOGRAPHICS | RADIOLOGY | RSNA JOURNALS ONLINE |
