Does Positron Emission Tomography Prevent Nontherapeutic Pulmonary Resections for Clinical Stage IA Lung Cancer?

Transcription

1 Does Positron Emission Tomography Prevent Nontherapeutic Pulmonary Resections for Clinical Stage IA Lung Cancer? Benjamin D. Kozower, MD, Bryan F. Meyers, MD, Carolyn E. Reed, MD, David R. Jones, MD, Paul A. Decker, MS, and Joe B. Putnam, Jr, MD Department of Surgery, University of Virginia Health System, Charlottesville, Virginia; Department of Surgery, Washington University School of Medicine, St. Louis, Missouri; Department of Cardiothoracic Surgery, Medical University of South Carolina, Charleston, South Carolina; Division of Biostatistics, Mayo Clinic, Rochester, Minnesota; and Department of Thoracic Surgery, Vanderbilt University Medical Center, Nashville, Tennessee Background. The American College of Surgeons Oncology Group (ACOSOG) Z0050 trial demonstrated that positron emission tomography (PET) prevents nontherapeutic thoracotomies in a substantial fraction of patients positive predictive value for advanced disease was only 33.3% (9 of 27; 95% CI, 16.5 to 54.0). The negative with known or suspected non-small cell lung cancer predictive value of PET to predict benign lesions was (NSCLC). However, the benefit of PET in clinical stageonly 57% (16 of 28; 95% CI, 37.2 to 75.5). Thus, 43% (12 of IA patients has been questioned due to the lower prev-28alence of metastases and poor ability to discriminate primary lesion actually had cancer, and all of these had 95% CI, 24.5 to 62.8) of patients with a PET negative benign from malignant lung lesions. This study evaluates whether PET prevents nontherapeutic pulmonary Conclusions. In clinical stage IA lung cancer patients, resectable disease (stages IA to IIB). resections in clinical stage IA patients by finding PET prevents nontherapeutic pulmonary resections less advanced disease or by declaring a nodule as benign. than 10% of the time. If a strategy of no surgery and serial Methods. We reanalyzed all patients with clinical stage computed tomographic scans is chosen for PET negative IA NSCLC from ACOSOG Z0050. The clinical, PET, andlesions, over 40% of patients with NSCLC will have pathologic stages were compared for this prospective surgery delayed. A prospective trial comparing PET versus resection for clinical stage IA lesions would clarify cohort. Results. One hundred twenty- two clinical stage IA the value of PET for these patients. patients were evaluated and 78.7% (96 of 122; 95% confidence interval [CI], 70.4 to 85.6) were eventually shown to have cancer. PET correctly showed 7.4% (9 of 122; 95% CI, 3.4 to 13.5) of patients to have advanced disease (stages IIIA to IV). However, due to a high false positive rate, the (Ann Thorac Surg 2008;85: ) 2008 by The Society of Thoracic Surgeons The American Cancer Society projects a lung cancer The increased use of computed tomographic (CT) scanning has resulted in a decrease in the median tumor size of incidence of 213,380 new cases for the year 2007 [1]. Lung cancer is the leading cause of cancer death world-resectewide and non-small cell lung cancer (NSCLC) accounts [3, 4]. In addition, the International Early Lung Cancer lung cancer and a shift toward earlier stage disease for 80% of newly diagnosed lung cancer cases. TheAction Program recently published follow-up on over optimal management of patients with NSCLC depends 31,000 asymptomatic people at risk for lung cancer who on accurate staging. The American College of Surgeons underwent screening chest CT [5]. The 10-year survival rate Oncology Group (ACOSOG) performed a multicenter, for stage I patients who underwent surgical resection was prospective, cohort trial (Z0050) to evaluate the utility of92%. Encouraging and provocative results like these are positron emission tomography (PET) in staging potentially operable NSCLC [2]. The authors concluded that by However, the American College of Chest Physicians does likely to increase the number of screening chest CT scans. identifying advanced disease (stages IIIA, IIIB, and IV) ornot recommend screening for lung cancer, even in high risk benign lesions, PET could potentially avoid unnecessary thoracotomies in 20% of patients. Accepted for publication Jan 4, Presented at the Fifty-fourth Annual Meeting of the Southern Thoracic Surgical Association, Bonita Springs, FL, Nov 7 10, Address correspondence to Dr Kozower, University of Virginia Health System, General Thoracic Surgery, PO Box , Charlottesville, VA ; bdk8g@virginia.edu. patients, outside of clinical trials. Many surgeons do not agree with this practice and the results of prospective screening trials using low dose CT scans are in progress [6, 7]. Although PET has become widely accepted as a routine staging test for NSCLC, the true benefit of PET for these small, early stage cancers remains unknown [2, 8, 9]. This study evaluates whether PET prevents nontherapeutic pulmonary resections in clinical stage IA patients by finding advanced disease or declaring a nodule benign by The Society of Thoracic Surgeons /08/$34.00 Published by Elsevier Inc doi: /j.athoracsur

2 Ann Thorac Surg KOZOWER ET AL 2008;85: PET AND STAGE IA LUNG CANCER Material and Methods This is a subset analysis examining patients with clinical stage IA NSCLC enrolled in the ACOSOG Z0050 trial. The detailed protocol for the Z0050 trial has been reported previously [2]. Briefly, patients with a biopsy-proven NSCLC or a suspicious nodule who were surgical candidates (stage I, II, or IIIA disease) after standard imaging procedures were eligible. The standard imaging procedures included CT of the chest and upper abdomen, bone scintigraphy, and contrast-enhanced CT or magnetic resonance imaging (MRI) of the brain. A prospective cohort of 303 enrolled patients then underwent PET scanning with the glucose analogue 18F-fluoro-2-deoxy-D-glucose (FDG). The PET was considered positive if the FDG uptake was slightly or substantially greater than blood pool activity. The protocol required that mediastinal abnormalities detected by PET be histologically confirmed. Suspected systemic metastases detected by PET also required confirmatory evaluation. The University of Virginia Institutional Review Board approved this study and waived the need to obtain individual consent (December 20, 2007). This subset analysis evaluated the 125 of 303 enrolled patients who were clinical stage IA and compared the clinical, FDG-PET, and pathologic stages. The radiology, pathology, and operative reports were reviewed to verify the results. Data were analyzed using SAS version 9.1 (SAS Institute Inc., Cary, NC). Data were summarized using point estimates and 95% exact binomial confidence intervals. Results This subset analysis includes 122 of the 125 clinical stage IA patients from the Z0050 trial. Two patients with pathologically confirmed mediastinal nodal disease were excluded because their official CT reports were suspicious for nodal disease (nodes 1 cm), making them clinical stage IIIA. The third patient was excluded because the PET evaluation of the lung lesion was negative, no biopsies were performed to verify the diagnosis, and the patient refused any follow-up. Table 1. Characteristics of Clinical Stage IA Patients (n 122) Characteristic No. (%) Age (years) ; Male 51 (41.8) Performance status 0 97 (79.5) 1 25 (20.5) Tumor location LUL 35 (28.7) RUL 34 (27.9) RLL 25 (20.5) LLL 19 (15.6) RML 9 (7.4) Values are n (%) unless indicated by mean SD, median, and range. LLL left lower lobe; LUL left upper lobe; RLL right lower lobe; RML right middle lobe; RUL right upper lobe. Table 2. Final Histology of Clinical Stage IA Patients The baseline characteristics of age, sex, performance status, and tumor location are listed in Table 1. The final histologies are listed in Table 2. Benign disease was found in 26 of 122 (21.3%) clinical stage IA patients and adenocarcinoma was the most common tumor (37.7%, 46 of 122). The primary way that PET can prevent a nontherapeutic pulmonary resection is by identifying metastatic disease (mediastinal lymph node involvement, or systemic metastasis). When examining the ability of PET to identify metastases, PET correctly labeled 7.4% (9 of 122; 95% CI, 3.4 to 13.5) of patients as having advanced disease (stages IIIA to IV). However, the positive predictive value of PET for advanced disease (stages IIIA to IV) was only 33.3% (9 of 27; 95% CI, 16.5 to 54.0). The high rate of false positives was attributed to the following: granulomatous disease (6), second primary tumor (3), adrenal adenoma (2), esophagitis (2), postoperative inflammation (1), and unknown (4). Pathologic analysis demonstrated that the prevalence of mediastinal nodal disease and systemic metastases in this analysis of 122 patients with clinical stage IA disease was 9% (11 of 122) and 1.6% (2 of 122), respectively. The other way that PET can prevent a nontherapeutic pulmonary resection is by correctly declaring a nodule benign and therefore suggesting a nonoperative strategy. The negative predictive value of PET to predict benign lesions was only 57% (16 of 28; 95% CI, 37.2 to 75.5). Of the 12 patients with a primary cancer despite a benign PET reading (a false negative PET), all had resectable NSCLC (stages IA to IIB). Mean tumor size for this group was cm with a range of 0.7 to 2.5 cm. Reasons for a false negative PET included the following: size less than 1 cm (2), and bronchioalveolar histology (4). Causes for the other six false negatives are unknown. There were no carcinoid tumors in this group of 12 patients. Importantly, the Z0050 trial enrolled patients with suspected bronchioalveolar carcinoma (BAC) and subcentimeter nodules, which may overestimate the false negative rate of PET. Six of these patients were included in the false negative PET group. If these six patients are excluded, as many physicians would not obtain a PET scan for these indications, the false negative rate is reduced from 43% to 27% (16 of 22). Comment Histology n (%) Benign 26 (21.3) Adenocarcinoma 46 (37.7) Squamous 25 (20.5) Bronchioalveolar 8 (6.6) Epidermoid 4 (3.3) Large cell 2 (1.6) Small cell 1 (0.8) Other 10 (8.2) 1167 It is widely accepted that PET improves the detection rate of local and distant metastases and can prevent non- GENERAL THORACIC

3 1168 KOZOWER ET AL Ann Thorac Surg PET AND STAGE IA LUNG CANCER 2008;85: therapeutic pulmonary resections for people with metastatic NSCLC [8, 10 12]. However, all of these studies include a heterogeneous group of patients with potentially resectable NSCLC, including stages IA through IIIA. The utility of PET to detect metastatic disease depends on the accuracy of the test and the true prevalence of metastatic disease in the population. The prevalences of mediastinal nodal disease and systemic metastases in this subset analysis were 9% (11 of 122) and 1.6% (2 of 122), respectively. This is an extremely important issue as the increased use of CT scanning has resulted in a decrease in the median tumor size of resected NSCLC and a shift toward earlier clinical stage at the time of presentation [3, 4]. Therefore, the ability of PET to prevent nontherapeutic pulmonary resections for patients with clinical stage IA disease must be challenged. A single center retrospective analysis from the Weill- Cornell Medical Center questioned the benefit of PET in the diagnosis and staging of tumors 2 cm in size or less [13]. This retrospective report on 64 patients concluded that PET had no significant benefit for this group of patients. The authors stated that PET was not useful due to the low prevalence of mediastinal nodal involvement and systemic metastases for these small tumors. The rate of mediastinal nodal involvement in this study was 9.4% (6 of 65) and was very similar to our analysis (9%, 11 of 122). The only randomized controlled trial to evaluate the use of PET in early stage (stage I and II) NSCLC is from Australia and concluded that PET did not significantly reduce the number of nontherapeutic thoracotomies [14]. However, there were two major issues with the Australian trial that make it difficult to interpret. First, all patients enrolled in the trial had a biopsy proven NSCLC. This is not standard practice in the United States. Many patients are brought to the operating room with an unbiopsied lesion suspicious for malignancy and more than 20% of them prove to be benign [2]. Therefore, the Australian analysis occurs in a clinical setting that reduces the potential utility of PET because PET cannot prevent nontherapeutic resections for benign lesions. Second, the majority of patients with ipsilateral mediastinal lymph node involvement (stage IIIA) in the United States are not offered primary surgery. Stage IIIA disease has a five-year survival rate of 15% and patients are usually managed with induction therapy followed by surgery or definitive chemotherapy and(or) radiation [15]. Because this study protocol was to operate on patients with known stage IIIA disease, only patients with contralateral mediastinal disease and distant metastatic disease (stages IIIB and IV) counted as nontherapeutic thoracotomies. These factors significantly underestimate the ability of PET to alter management compared with the anticipated ability in the United States. The high false positive rate of PET for clinical stage IA patients is concerning. Although most trials evaluating the role of PET to prevent nontherapeutic pulmonary resections caution about the relatively high false positive rate, clinicians do not always obtain a tissue biopsy to confirm the diagnosis before declaring a patient inoperable due to advanced stage [8, 14]. There is also a need to determine the economic costs and benefits of PET as a preoperative staging tool for early, stage IA NSCLC. A cost analysis of the PLUS trial (positron emission tomography in the preoperative assessment of patients with suspected non-small cell lung cancer), a randomized controlled trial evaluating PET in potentially operable patients, suggested a significant cost savings for using PET in potentially resectable patients with NSCLC. However, these studies have not looked specifically at stage IA NSCLC patients. This is an important distinction because small tumors and microscopic nodal metastases will not be FDG avid [16, 17]. Because these smaller tumors have a lower likelihood for mediastinal or distant metastases, there will be a higher burden of potentially expensive and bothersome confirmatory studies to verify false positive PET scans in this population compared with a cohort with a higher prevalence of metastatic disease. Several reports have demonstrated that PET is cost effective in assessing solitary pulmonary nodules [18, 19]. However, there is a selection bias for patients who are seen by a thoracic surgeon and included in the Z0050 trial for biopsy proven or highly suspicious NSCLC. It is unknown how many of these patients had already been observed and how many were sent for surgical consideration because of growth in the suspicious lesion. It is difficult to apply cost-effectiveness studies of solitary pulmonary nodules in the general population to this highly selected group of surgically referred patients. The false negative rate of PET for declaring nodules benign in our study was 43%. This high false negative rate makes it risky to follow, rather than resect, based on the negative PET result alone. Looking at our results another way, the risk of a benign resection before PET was 21.3% (26 of 122). This rate can be reduced to 10.6% (10 of 94) using PET if patients with a negative PET are followed. However, to achieve this risk reduction one would need to delay (watch and wait) on 26 patients, 12 of whom have NSCLC despite a benign PET result. This study has several limitations. First, it is an exploratory subset analysis and the Z0050 trial was not designed to look specifically at the use of PET in stage IA NSCLC. Second, Z0050 was performed using standard FDG-PET rather than integrated PET-CT. The integration of PET with CT (integrated PET-CT) is more accurate than PET alone [20, 21]. In addition, the PET was considered positive if the FDG uptake was slightly or substantially greater than blood pool activity per the Z0050 protocol. Standardized uptake values were not included with the majority of reports in this trial. Third, there is a possible selection bias among potential clinical stage IA patients who were enrolled in the trial that may not generalize to all clinical stage IA NSCLC patients. It is unknown how many of these patients were being followed for pulmonary nodules or how many had an increase in nodule size. Fourth, the Z0050 trial enrolled patients with suspected BAC and subcentimeter nodules. Many physicians would not obtain a PET scan for these indications and this may overestimate the false negative rate for PET in clinical stage IA disease. Finally, the impact of PET will vary depending on the clinical practice of the surgeon. For example, most surgeons will not perform cervical mediastinoscopy on a clinical stage IA

4 Ann Thorac Surg KOZOWER ET AL 2008;85: PET AND STAGE IA LUNG CANCER patient with a negative PET in the mediastinum [22]. However, if omitting the PET scan will then lead to routine mediastinoscopy in these early stage patients, the cost savings of omitting PET might disappear or even reverse. In conclusion, this subset analysis of a prospective, multicenter trial demonstrates that routine PET prevents nontherapeutic pulmonary resections in 7.4% of patients with clinical stage IA NSCLC by identifying metastatic disease. This is significantly lower than the prevention of 20% of nontherapeutic pulmonary resections suggested by the Z0050 and the PLUS trials, both of which evaluated all potentially resectable patients (stage IA-IIIA) [2, 8]. If a strategy of deferring surgery and offering serial CT scans is chosen for PET negative lesions, over 40% of patients with NSCLC will have their surgery delayed. A randomized controlled trial would provide the most reliable and convincing answer to determine the utility of PET for stage IA NSCLC. It would provide broad comparability of groups except for the diagnostic PET being assessed and ensure a nonbiased assessment of the value of the test with respect to the clinically relevant outcomes [23]. We believe that PET needs to be used carefully in these patients and that metastatic findings need tissue confirmation. For PET negative, but clinically suspicious lesions, physicians need to use their clinical judgment in deciding whether to follow these patients, perform a transthoracic needle biopsy, or perform an excisional biopsy. References American Cancer Society. Cancer facts & figures. Atlanta: American Cancer Society; Reed CE, Harpole DH, Posther KE, et al. Results of the American College of Surgeons Oncology Group Z0050 trial: the utility of positron emission tomography in staging potentially operable non-small cell lung cancer. J Thorac Cardiovasc Surg 2003;126: Okada M, Nishio W, Sakamoto T, et al. Evolution of surgical outcomes for nonsmall cell lung cancer: Time trends in 1465 consecutive patients undergoing complete resection. Ann Thorac Surg 2004;77: MacMahon H, Austin JH, Gamsu G, et al. Guidelines for management of small pulmonary nodules detected on CT scans: a statement from the Fleischner Society. Radiology 2005;237: International Early Lung Cancer Action Program, Investigators, Henschke CI, Yankelevitz DF, et al. Survival of patients with stage I lung cancer detected on CT screening. N Engl J Med 2006;355: Bach PB, Niewoehner DE, Black WC, American College of Chest Physicians. Screening for lung cancer: the guidelines. Chest 2003;123(1 suppl):83s 8S. 7. Bach PB, Silvestri GA, Hanger M, Jett JR, American College of Chest Physicians. Screening for lung cancer: ACCP evidence-based clinical practice guidelines (2nd edition). Chest 2007;132(3 suppl):69s 77S. 8. van Tinteren H, Hoekstra OS, Smit EF, et al. Effectiveness of positron emission tomography in the preoperative assessment of patients with suspected non-small-cell lung cancer: The PLUS multicentre randomised trial. Lancet 2002;359: Silvestri GA, Tanoue LT, Margolis ML, Barker J, Detterbeck F, American College of Chest Physicians. The noninvasive staging of non-small cell lung cancer: The guidelines. Chest 2003;123:147S 56S. 10. Pieterman RM, van Putten JW, Meuzelaar JJ, et al. Preoperative staging of non-small-cell lung cancer with positronemission tomography. N Engl J Med 2000;343: Schrevens L, Lorent N, Dooms C, Vansteenkiste J. The role of PET scan in diagnosis, staging, and management of non-small cell lung cancer. Oncologist 2004;9: Rankin SC. Staging of non-small cell lung cancer (NSCLC). Cancer Imaging 2006;6: Port JL, Andrade RS, Levin MA, et al. Positron emission tomographic scanning in the diagnosis and staging of nonsmall cell lung cancer 2 cm in size or less. J Thorac Cardiovasc Surg 2005;130: Viney RC, Boyer MJ, King MT, et al. Randomized controlled trial of the role of positron emission tomography in the management of stage I and II non-small-cell lung cancer. J Clin Oncol 2004;22: Mountain CF. Revisions in the international system for staging lung cancer. Chest 1997;111: Nomori H, Watanabe K, Ohtsuka T, et al. Fluorine 18-tagged fluorodeoxyglucose positron emission tomographic scanning to predict lymph node metastasis, invasiveness, or both, in clinical T1 N0 M0 lung adenocarcinoma. J Thorac Cardiovasc Surg 2004;128: Nomori H, Watanabe K, Ohtsuka T, Naruke T, Suemasu K, Uno K. The size of metastatic foci and lymph nodes yielding false-negative and false-positive lymph node staging with positron emission tomography in patients with lung cancer. J Thorac Cardiovasc Surg 2004;127: Keith CJ, Miles KA, Griffiths MR, Wong D, Pitman AG, Hicks RJ. Solitary pulmonary nodules: accuracy and costeffectiveness of sodium iodide FDG-PET using Australian data. Eur J Nucl Med Mol Imaging 2002;29: Gambhir SS, Shepherd JE, Shah BD, et al. Analytical decision model for the cost-effective management of solitary pulmonary nodules. J Clin Oncol 1998;16: Lardinois D, Weder W, Hany TF, et al. Staging of non-smallcell lung cancer with integrated positron-emission tomography and computed tomography. N Engl J Med 2003;348: Cerfolio RJ, Ojha B, Bryant AS, Raghuveer V, Mountz JM, Bartolucci AA. The accuracy of integrated PET-CT compared with dedicated PET alone for the staging of patients with nonsmall cell lung cancer. Ann Thorac Surg 2004;78: Meyers BF, Haddad F, Siegel BA, et al. Cost-effectiveness of routine mediastinoscopy in computed tomography- and positron emission tomography-screened patients with stage I lung cancer. J Thorac Cardiovasc Surg 2006;131: Van Tinteren H, Hoekstra OS, Boers M. Do we need randomised trials to evaluate diagnostic procedures? Eur J Nucl Med Mol Imaging 2004;31: GENERAL THORACIC DISCUSSION DR MARK J. KRASNA (Towson, MD): Dr Merrill, Dr Fiore, members and guests. I want to congratulate Dr Kozower on an excellent presentation and thank him and his colleagues for sending me the manuscript in advance. This paper represents a subset analysis of the ACOSOG [American College of Surgeons Oncology Group] Z0050 trial studying PET [positron emission tomography] scanning in non-small cell lung cancer. The trial included biopsy proven lung cancers or patients with a suspicious nodule who were deemed to be surgical candidates after standard radiographic imaging. The goal was to determine how often could they avoid unnecessary surgery. Of the 303 patients enrolled, the study looked at 125 who were clinical stage IA; 122 of these were actually eligible. Twenty-six lesions turned out to be benign, leaving 96 cancers in this trial to be analyzed. PET

5 1170 KOZOWER ET AL Ann Thorac Surg PET AND STAGE IA LUNG CANCER 2008;85: correctly identified only 7.4% of these patients as having advanced disease; that is, stage IIIA or stage IV. The positive predictive value was only 33%. This high false positive rate means more patients not being offered surgical resection or, alternatively, more patients being offered neoadjuvant therapy when in fact they have early stage disease; less problematic but not exactly an accepted mode of therapy in stage III non-small lung cancer. The negative predictive factor was also low, only 57%. Although all 12 of these cancers that were missed by PET were eventually resectable, as you noted, these patients had a delay in therapy. And finally, only four of these were due to bronchoalveolar histology, which is assumed to be one of the most common problems. This high false negative rate would mean that patients were getting surgery for advanced disease or patients who had missed cancers having delays in therapy long term. This is indeed sobering data given such a homogeneous group of patients. Although there were only 9% of the patients who were missed N2 disease, in fact, the expected five-year survival for those patients would probably be any place from 9% to, at best, 20%. Lastly, this study included only those patients who were seen by a thoracic surgeon. This is important, because unlike the Australian study which showed no benefit in PET scan in early stage lung cancer, that study did biopsies first and then did PET scans. In the United States, we usually would do the PET scan first to rule out metastatic disease or identify mediastinal disease and then do the biopsy. The Australian approach was also to operate on patients with IIIA disease anyway, which you probably would not do here in the United States given the widespread use of neoadjuvant therapy. Finally, as you mentioned, the analysis may be underpowered since this subset study was not in the initial design. I have two questions for you then. First, how will you and your co-authors use PET in the future for clinical stage I lesions? Will it be used routinely, in fact, or will you do the PET only when you have a biopsy? And second, if you use PET scan routinely, how will you use minimally invasive surgical mediastinal lymph node staging such as EUS [endoscopic ultrasonography], or EBUS [endobronchial ultrasound], or mediastinoscopy? Are you going to use that routinely given these results or only when the PET scan is positive? Again, thank you for an excellent presentation and I thank the Association for the privilege of discussing this paper. DR KOZOWER: Thank you very much for your comments and your two questions. The first question is how will we use PET going forward? I think the purpose of this study was to question the routine use of PET for early stage lung cancer. The group from Cornell showed a very similar finding when they looked at their tumors less than 2 cm in size. I think the only way to answer this question is to perform a prospective trial and include an analysis of cost effectiveness. DR KRASNA: So, next week on Monday, then what are you going to do at UVA for a clinical stage IA? DR KOZOWER: Typically, for a small ( 1.5 cm), peripheral, TIN0 lesion I do not obtain a PET scan. However, many patients have already had it done when they come to their clinic appointment. Your second question is important and asks about the influence of PET on other mediastinal staging strategies. Bryan Meyers decision analysis published in JTCVS (Journal of Thoracic Cardiovascular Surgery) demonstrated that if you have a clinical stage IA lung cancer by CT [computed tomography] and PET, the rate of occult N2 disease is quite low and mediastinoscopy didn t offer much. I don t think the data would be very different for EUS or EBUS. The question then becomes, is it better to stage the mediastinum with EBUS and EUS or PET? I think it will be very difficult to show that it s cost effective to perform all of these tests when the prevalence of occult N2 disease is low. DR ROBERT J. CERFOLIO (Birmingham, AL): I have so many questions now I don t know where to start. I thought you were doing great until you told me you wouldn t do a PET for patients with stage IA disease. I would understand it if you looked at your data, but what about those 7% of people that you showed you would have helped, number one? Number two, how did you define false positives? Did you just use an arbitrary cutoff of 2.5? Because if you did, then I would say that ain t the PET s fault, that is your fault for not reading it right. Did you use a ratio of the max SUV of the primary tumor to the lymph node s max SUV? Did you use an absolute value for the max SUV of the lymph node? DR KOZOWER: Dr Cerfolio, that is an excellent point. The Z0050 trial performed their PET scans using the glucose analogue 18F-fluoro-2-deoxy-D-glucose (FDG). The PET was considered positive if the FDG uptake was slightly or substantially greater than blood pool activity. Unfortunately, many of the reports did not give SUV [standardized uptake] values and I agree with you that the quality of the scans has improved. This is a limitation of the Z0050 trial and it is a limitation of this subset analysis. DR CERFOLIO: That is the problem. So that falsely elevates your false positive rate through the roof, because we all know that granulomatous lymph nodes are going to have max SUVs of 3, 4, 5, and 6. And finally, you talk about how it hurt 40% of patients because they didn t get surgery. Do you have any data that if a patient who has a cancer that has no FDG avidity is harmed by waiting a year or maybe even two prior to resection? Is there data that that wait causes the stage to be higher after resection than it would have been two years earlier? Because we have data that it may not change at all since it grows so slowly. Do you have any idea that you really hurt those patients by not operating on them and maybe waiting to operate three, six, nine, 12 months, or even 24 months later when the tumor grew? DR KOZOWER: It is a fair question and I don t have that data. It would be a good idea to try and look at some retrospective data but a prospective study would be impossible. DR CERFOLIO: It would be difficult to do it prospectively but retrospectively you could show, as we can, that patients who had a negative PET who got resected after a period of follow-up because of the nodule s growth were all node negative and the T status was unchanged. And then finally, you say that it missed some N2 disease and that that would be bad, but again do you have data that patients who have integrated PET-CT N2 negative disease and CT N2 negative disease do any better with neoadjuvant therapy prior to resection as opposed to resection followed by adjuvant therapy? We have some data on the favorable survival of these patients and we will be presenting it at the STS [Society of Thoracic Surgeons]. My point is their survival with integrated PET/CT N2 disease that is resected is pretty good. Thus, in conclusion, maybe you should PET that patient with your assumed clinically stage IA disease because it may be stage IV, so PET isn t so.