| HOME | HELP | CONTACT US | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Click here to read this article as a CME course
Lung Cancer |
Departments of aMedical Oncology and bRadiology and cInstitute of Hygiene and Epidemiology, University Hospital of Udine, Udine, Italy; dUnit of Occupational Health and Departments of eRadiology, fOncology, and gSurgery, San Polo Hospital, Monfalcone, Italy; hMedical Oncology A, Disease Management Team - Lung Cancer, National Institute for Cancer Research, Genoa, Italy
Key Words. Occupational exposure • Asbestos • Screening • Spiral computed tomography • Lung cancer Malignant pleural mesothelioma
Correspondence: Gianpiero Fasola, M.D., Department of Medical Oncology, University Hospital of Udine, P.le S. M. Misericordia 15, 33100 Udine, Italy. Telephone: 39-0432-552-750; Fax: 39-0432-552-751; e-mail: fasola.gianpiero{at}aoud.sanita.fvg.it
Received July 9, 2007; accepted for publication August 21, 2007.
Disclosure: P.C. received an honorarium for his contribution to the study. No other potential conflicts of interest were reported by the authors, planners, reviewers, or staff managers of this article.
Access and take the CME test online and receive 1 AMA PRA Category 1 CreditTM at CME.TheOncologist.com
Learning Objectives
Top
Learning Objectives
Abstract
Introduction
Participants and Methods
Results
Discussion
Conclusion
Acknowledgments
References
After completing this course, the reader will be able to:
|
ABSTRACT |
|---|
|
TopLearning ObjectivesAbstractIntroductionParticipants and MethodsResultsDiscussionConclusionAcknowledgmentsReferences |
|---|
Methods. Between February 2002 and October 2003, 1,045 volunteers already enrolled in a surveillance program for asbestos-exposed workers and former workers were recruited. The main eligibility criteria were: written informed consent, definite exposure to asbestos, age 40–75, no prior cancer or severe concomitant conditions, no chest CT scan in the past 2 years. A smoking history was not required. After a structured interview, chest X-ray (CXR) and LDCT were performed. Participants with negative examinations were assigned to annual LDCT. Participants with positive findings received high-resolution CT and additional diagnostic workup as appropriate.
Results. Baseline characteristics of the screened population were: median asbestos exposure time, 30 years; median age, 58; median pack-years in smokers/former smokers, 18.5. Thirty-four percent had never smoked. On LDCT, 834 noncalcified nodules were identified in 44% of participants, versus 43 nodules in 4% on CXR. Pleural abnormalities were observed in 44% and 70% of participants by CXR and LDCT, respectively. Overall, LDCT identified nine cases of non-small cell lung cancer—eight stage I, one stage IIA—and one thymic carcinoid, corresponding to 1% of the enrolled population. All cases were radically treated. None had been detected by CXR. No pleural mesothelioma was diagnosed. There were 11 false-positive results.
Conclusions. Our findings first suggest that LDCT may be at least as useful in asbestos workers as in heavy smokers for the early diagnosis of lung cancer; this benefit is evident even in a poor-risk population, with low rates of smoking prevalence and a previous history of radiological surveillance.
The role of spiral tomography in screening for pleural mesothelioma remains uncertain.
|
INTRODUCTION |
|---|
|
TopLearning ObjectivesAbstractIntroductionParticipants and MethodsResultsDiscussionConclusionAcknowledgmentsReferences |
|---|
The most important prognostic factor in both malignancies is disease stage. Long-term survival may be >80% in early-stage non-small cell lung cancer (NSCLC), but these rates fall to <5% in more advanced cases. Also, selected patients with early MPM may experience a better outcome with aggressive multimodal treatments. In some centers, this approach has led to median survival times of about 2 years [3, 4].
Overall, both lung cancer (worldwide) and pleural mesothelioma (in selected areas) would be suitable targets for screening: they represent a significant public health issue, have a long asymptomatic natural history, and may benefit from early therapeutic intervention [5].
In the 1970s, randomized trials of lung cancer screening with chest X-ray (CXR) and sputum cytology failed to show a significant reduction in disease-specific mortality among high-risk male smokers [6–9].
Interest in screening was renewed at the turn of this century with the introduction of low-dose spiral computed tomography (LDCT). Several studies in the past 10 years have reported encouraging results, showing that LDCT is four times as sensitive as CXR in detecting small, potentially resectable lung cancers [10], whereas its mortality benefit has not yet been fully elucidated.
If feasibility trials of LDCT screening in heavy smokers have been widely conducted, data are missing about its worth in other high-risk populations, such as asbestos-exposed individuals. Yet, the causal relationship between asbestos exposure and both malignant mesothelioma [11] and lung cancer [12, 13] is well established; cigarette smoking and asbestos may also exert a joint, less than multiplicative but supra-additive effect on lung cancer risk [14].
In the coastal area of Trieste and Monfalcone in northeastern Italy, standardized incidence rate ratios of malignant mesothelioma, as compared with those from Italy as a whole, are approximately ninefold and fivefold higher in men and women, respectively [15–17]. Most of the cases are observed among shipyard workers and their wives, with a well-known history of asbestos exposure. The corresponding ratio for lung cancer is about 1.3 for both sexes, and the population fraction of lung cancer attributable to possible or definite exposure to asbestos has been estimated to be 20% (95% confidence interval [CI], 11.5%–28.5%) [18].
Considering the relevance of asbestos-related malignancies in that area and the current lack of evidence about LDCT screening in this risk setting, we designed a prospective, nonrandomized trial to evaluate the feasibility of baseline and annual repeat screening with LDCT among asbestos-exposed workers and former workers. Results are reported for the prevalence phase of the study.
|
PARTICIPANTS AND METHODS |
|---|
|
TopLearning ObjectivesAbstractIntroductionParticipants and MethodsResultsDiscussionConclusionAcknowledgmentsReferences |
|---|
A surveillance program for asbestos-exposed workers and former workers has been active since 1994 at the Unit of Occupational Health in Monfalcone. Enrolled participants—representing an estimated one third of all the asbestos-exposed workers of the region—annually undergo a physical examination, respiratory function tests, and a CXR. Program participants were considered possible but not exclusive candidates for our study.
At enrolment, eligible individuals underwent a structured interview about medical and family history, demographics, residential history, military service, diet, smoking, and alcohol consumption. Occupational history was recorded in detail: participants were asked about place(s) of employment, exact tasks, type and duration of asbestos exposure, use of protective equipment, and cleaning of contaminated clothing.
The study protocol was approved by the local ethical committee.
Screening Procedures
Blood samples were collected from each participant and stored for future analyses. Respiratory function tests were also performed at baseline. Participants underwent CXR and LDCT. Posterior–anterior and lateral chest radiographs were obtained according to standard protocols. Helical LDCT scans were performed with a GE Medical System scanner (GE Medical Systems, Tokyo, Japan) at 120 kV, 40 mA, 0.8 second/rotation, and 2:1 pitch with a slice thickness of 5 mm. Scans were obtained from the level of the apex to the diaphragm in a single 20-second breath-hold at end inspiration, after 1 minute of hyperventilation. Image reconstruction was performed with a standard algorithm at intervals of 5 mm. All LDCT images were independently reviewed by two radiologists. Their findings were individually recorded and then discussed, and the consensus findings were documented.
When nodules were identified, the defined characteristics of each were recorded: size, location, benign calcifications, shape, and edge. Matching of LDCT findings with CXR was checked for each nodule. Radiologists were asked to classify abnormalities other than nodules into the following categories: pleural thickening/plaque, pleural effusion, parenchymal focal opacity, endobronchial lesion, fibrosis/scar, bone/soft tissues lesion, cardiac abnormality, emphysema/chronic obstructive broncopneumopathy, other.
Positive baseline exams (Fig. 1 and Fig. 2) were defined as follows: noncalcified nodules (NCNs), calcified nodules >20 mm or with malignant pattern, and pleural thickening >10 mm or with scissural or circumferential involvement. Participants with negative baseline exams were proposed for annual repeat LDCT. Those with positive findings underwent high resolution CT (HRCT) and additional diagnostic workup as appropriate.
|
|
If malignant and resectable disease was diagnosed, radical surgery on the primary lesion was coupled with mediastinal lymph-node dissection and labeling of all lymph-node locations. All cytological and histological findings from any biopsy and surgical procedure were documented.
Statistical Considerations
The baseline phase of the study was designed to show a 9% higher rate for CT-diagnosed tumors compared with annual CXR, assuming a lung cancer probability in the study group at least as high as in the ELCAP study [19]. Using a one-sided, 0.05-level test, the enrolment of at least 832 subjects was required to obtain a statistical power of 80%.
|
RESULTS |
|---|
|
TopLearning ObjectivesAbstractIntroductionParticipants and MethodsResultsDiscussionConclusionAcknowledgmentsReferences |
|---|
Table 1 shows the results of baseline CXR and LDCT. CXR identified 43 nodules. Thirty-four participants had one nodule, three had two nodules, and only one had three nodules. LDCT detected 834 NCNs (750 were <5 mm, 73 were 6–10 mm, and 11 were >10 mm) and 81 calcified nodules. Two hundred sixty participants had one NCN, 187 had 2–5, and 13 had 6–9. Overall, 460 subjects had at least one NCN. The right lung was involved more frequently than the left. Four hundred seventy-seven pleural abnormalities were detected in 44% of the participants with CXR, versus 880 in 70% of the participants with LDCT. No pleural effusion was identified with either technique.
|
|
|
BAC, adenocarcinomas, and the NSCLC-NOS were treated with lobectomy, except for one case of BAC, in which surgery was limited to wedge resection; the carcinosarcoma was treated with right pneumonectomy, while the thymic carcinoid was resected in sternotomy en bloc with pericardium. All cases, except for one NSCLC case, were stage I; the remaining one was stage IIA.
Overall LDCT identified 10 thoracic malignancies (0.96% of the enrolled participants, 1.9% of recalls), nine lung cancers, and one thymic carcinoid. None of the malignancies had been identified by initial CXR.
Because of the small number of lesions and the high level of homogeneity in asbestos exposure among study participants, no association could be found between the 10 screen-diagnosed malignancies and age, smoking, duration of asbestos exposure, and time since first exposure.
NCNs according to LDCT and CXR are displayed in Table 3. NCNs were not associated with age or with asbestos exposure, a somehow expected result because these factors are risk indicators for malignancy and most of the nodules were benign. Notably, the prevalence odds ratio for at least 25 pack-years is 1.49 (95% CI, 1.06–2.09).
|
|
|
DISCUSSION |
|---|
|
TopLearning ObjectivesAbstractIntroductionParticipants and MethodsResultsDiscussionConclusionAcknowledgmentsReferences |
|---|
Few data are currently available with respect to the value of LDCT screening in other high-risk populations. In the prevalence phase of a study that screened 602 asbestos workers using both CXR and LDCT, five cases of lung cancer were detected—one stage II, two stage IIIB, and two stage IV. One case had been originally misinterpreted as negative. Overall, the study showed some inaccuracy in the diagnostic workup of suspicious findings and incongruity in their treatment [20]. More recently, two small studies were conducted, each including an asbestos-exposed cohort of <200 individuals [21, 22]. Notably, in all three cases, study participants were mostly smokers/former smokers (97%, 99%, and 100%, respectively). To date, no LDCT screening study has been reported among asbestos workers with a sample size comparable with that of the original ELCAP trial [19].
Our study has met its primary endpoint, showing that, in asbestos-exposed individuals, LDCT, as compared with CXR, significantly increases the likelihood of detecting small NCNs, and thus lung cancer, at an earlier, more curable stage. Indeed, we detected NCNs 19 times more frequently with LDCT than with CXR. None of the 10 CT-detected malignancies had been identified by initial CXR. All of them could be radically resected; no death or major complications occurred, confirming previous observations of a more favorable morbidity profile of early lung cancer resection, compared with resection of symptomatic, more advanced cases [23, 24].
The proportion of lung cancer cases was apparently only in the range of those from other screening trials (0.4%–2.7%) [10]. Study participants had already entered a surveillance program, and more than one third had undergone a CXR before enrolment; in other words, they represented a highly selected population, somewhat resembling an incidence rather than a prevalence group. In addition, they were a relatively low-risk cohort: the median age was 58, 34% had never smoked, and the median exposure for smokers and former smokers was 18.5 pack-years. Thus, our results compare very favorably with those of previous screening studies, whose rate of lung cancer detection among individuals with a median exposure of 40–45 pack-years was in the range of 1.1%–2.7% [19, 23–25].
As with any screening intervention, this encouraging outcome should be balanced against the possibility that aggregate harm to screened individuals may exceed the benefit conferred to those who have lung cancer detected [10]. In particular, harm may result from the high costs of screening and follow-up, and the morbidity associated with false-positive results. In our experience, the false-positive rate was about 1%; indeed, 52% of all invasive procedures were performed for benign conditions, a proportion that largely exceeds the 25% rate reported in other studies. This is mainly a result of our choice of using VATS for a definite pathological diagnosis of suspicious nodules. VATS was chosen for a number of reasons. Following a multidisciplinary evaluation, most nodules were found to be easier to manage with a surgical approach. We considered that, in many centers, the expertise needed for fine-needle aspiration biopsy (FNAB) of small and deeply located lesions is still unavailable, as well as techniques like positron-emission tomography. Besides, in a patient with a new, solitary, NCN >5 mm, surgical resection may be the ideal approach, as it is both diagnostic and therapeutic [26]. FNAB is obviously less invasive, but it can be, at best, diagnostic only. While its sensitivity for malignancy can be >60%, the false-negative rate is 3%–29%, and the incidence of pneumothorax is up to 30% [27]. Finally, the sensitivity of bronchoscopy for detecting a malignant process in a solitary pulmonary nodule <1.5 cm in diameter is only 10% [27].
Beyond the choice of the most appropriate diagnostic tool, the adoption of a less stringent diagnostic algorithm would have been of benefit. Notably, Henscke and colleagues recently updated the ELCAP original protocol [28], recognizing that, in modern CT screening for lung cancer, NCNs <5 mm detected at baseline do not justify immediate workup, but only annual repeat screening to determine interim growth. In our study, the rate of detection of NCNs corresponded to 80% of the enrolled population, whereas that of nodules >5 mm was only 8%. We estimated that the modified algorithm would have reduced the rate of unnecessary invasive procedures by >30%.
In any case, none of the participants suffered complications from the diagnostic procedures.
Our findings do not seem to support the use of LDCT in the early diagnosis of pleural mesothelioma, although some caution is needed when interpreting these data: the current MPM incidence rate in the study area is approximately 10 per 100,000 per year, corresponding to an absolute number of eight cases. Even if we assumed that the risk for MPM for individuals exposed to asbestos was much higher (e.g., 10 times) than in the general population, and that LDCT could screen a prevalence of 2 per 1,000 versus a prevalence of 0.5 per 1,000 with CXR, with an 
error of 0.05 and a power of 80%, the minimum sample size would be approximately 4,200 subjects. Indeed, considering the sample size and lifetime risk for MPM in asbestos-exposed individuals, less than one case was expected at baseline.
The study detected a very high number of pleural abnormalities, with LDCT showing a considerably higher accuracy rate than CXR. Pleural thickening and pleural plaques are commonly seen in patients without lung disease, but there is some controversy about their relationship with asbestos exposure indexes and with the risk for malignant evolution in the absence of asbestosis [29].
Annual repeat LDCT screening could provide useful information on the natural history and significance of asbestos-related pleural abnormalities. Still, differently from nodules, the malignant transformation process of pleural plaques probably requires many years and a long follow-up. From this perspective, CXR could involve a much lower expenditure, but at the cost of a significantly inferior diagnostic sensitivity; on the other hand, it is reasonable to expect that novel insights into the significance of pleural abnormalities may emerge from novel imaging techniques with higher resolution capabilities than 5-mm collimation CT. However, results of the 5-year follow-up are awaited: the finding of an association between early-detected pleural abnormalities and the subsequent risk for thoracic malignancies could be of remarkable interest in the long-standing debate on legal acknowledgment and reimbursement of exposure victims.
In any case, pleural abnormalities have been shown to have a significant relationship with age, smoking, and time since first asbestos exposure, but not with its duration. Possibly, even a short exposure increases the risk for plaques. Therefore, it becomes apparent that, should pleural abnormalities be predictors of any severe respiratory condition, their screening would be more appropriate in a cohort of aged heavy smokers, with a remote history of asbestos exposure, independently from its course.
|
CONCLUSION |
|---|
|
TopLearning ObjectivesAbstractIntroductionParticipants and MethodsResultsDiscussionConclusionAcknowledgmentsReferences |
|---|
The U.S. National Lung Screening Trial randomized 50,000 high-risk smokers to have either CXR or CT scans, with results expected by 2011, while in the Netherlands and Belgium, 16,000 individuals have been randomized to receive CT or usual care, with results due in 2016. Only these trials could establish whether a true reduction in lung cancer mortality results from screening with LDCT [30].
The obvious result of our study is that 1% of the participants did in fact have a previously unrecognized malignancy that was small enough to allow definitive treatment. This benefit was evident even in a poor-risk, low smoking, and already surveyed population. Should ongoing screening trials demonstrate a decrease in lung cancer mortality, the finding that LDCT may capture at least an equal proportion of lung cancers in asbestos workers as in heavy smokers could provide the rationale to perform well-designed validation trials in the former risk group as well.
|
ACKNOWLEDGMENTS |
|---|
|
TopLearning ObjectivesAbstractIntroductionParticipants and MethodsResultsDiscussionConclusionAcknowledgmentsReferences |
|---|
The authors gratefully acknowledge the excellent contribution of the study data manager, Dr. Marica Gaiardo. We would also like to thank Dr. Claudio Rieppi, Prof. Claudio Bianchi, Dr. Alessandro Brollo, Dr. Giovanni Pilati, Dr. Danilo Spazzapan, and Dr. Manuela Baccarin for their valuable support in conducting the study.
We thank the patients and their families for their confidence and collaboration.
G.F. and O.B. contributed equally to the study.
|
REFERENCES |
|---|
|
TopLearning ObjectivesAbstractIntroductionParticipants and MethodsResultsDiscussionConclusionAcknowledgmentsReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J. P. van Meerbeeck and G. Hillerdal Screening for Mesothelioma: More Harm than Good? Am. J. Respir. Crit. Care Med., October 15, 2008; 178(8): 781 - 782. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | CONTACT US | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| THE ONCOLOGIST | STEM CELLS | CME | ALPHAMED PRESS JOURNALS |
