Lung cancer association

Study objective: To identify factors associated with a misclassification of the true disease stage by comparing the differences between the clinical and pathologic stage of patients with early-stage non-small cell lung cancer (NSCLC).

Design: A prospective cohort study.

Setting: A multidisciplinary thoracic oncology clinic at a university-affiliated Veterans Affairs medical center.

Patient population: One hundred nine male veterans with clinical stage UII NSCLC who had undergone thoracotomy with systematic lymph node dissection.

Methods: Prospective data were collected on all patients between September 1997 and April 2002. Logistic regression analysis was used to establish the odds ratio (OR) for predictors of changes in stage.

Results: A stage misclassification was found in 35.8% of patients (39 of 109 patients) after thoracotomy with lymph node dissection, and all but one patient were upstaged. Unsuspected nodal involvement (N stage) resulted in the upstaging of 16.5% of the patients, a change in tumor stage (T stage) resulted in the upstaging of 13.8% of the patients, a change in both stages resulted in the upstaging of 2.7% of patients, and the designation of metastatic disease resulted in the upstaging of 1.9% of the patients. The rate of unsuspected mediastinal lymph node involvement (pathologic stage N2) was 8.3% (9 of 109 patients), despite negative mediastinoscopy findings. Complete anatomic resection was performed in all patients. Advanced disease was found in 8.3% of the patients (9 of 109 patients) [stage IIIB or IV]. Having the primary tumor in a lower lobe location was the only statistically significant factor associated with upstaging (OR, 3.56; 95% confidence interval, 1.4 to 9.1). The effect of location was robust alter controlling for tumor size and the prior performance of mediastinoscopy. Patient age, smoking history, weight loss, tumor size, and tumor histology were all found not to be associated with upstaging.

Conclusion: A lower lobe tumor location in patients with early-stage NSCLC appears to be associated With upstaging after surgery. We conclude that a tumor location in a lower lobe deserves special attention.

Key words: interdisciplinary communication: lung neoplasm; mediastinoscopy; neoplasm staging; prospective studies risk factors; tomography

Abbreviations: cTNM = clinical stage using the TNM system; MTOC = multidisciplinary thoracic oncology clinic; N stage = nodal involvement; NSCLC = non-small cell lung cancer; OR = odds ratio; PET = positron emission tomography; pTNM = pathologic stage using the TNM system; T stage = tumor stage

**********

Non-small cell lung cancer (NSCLC) is the most common histologic type of lung cancer and accounts for approximately 75 to 85% of all newly diagnosed lung neoplasms. (1) Once the tissue diagnosis has been established, the workup focuses on the most accurate assessment of the extent of the disease. (2) This is because surgical resection offers a promise for cure in those patients with early-stage lung cancer who do not show involvement of the mediastinal lymph nodes or distant disease. Even though the combined surgical and pathologic stage using the TNM system (pTNM) defines the disease most accurately, ascertaining the clinical stage using the TNM system (cTNM) is a crucial step in the decision to recommend that the patient undergo a potentially curative resection. The cTNM combines the interpretation of the patient's history, physical examination findings, laboratory data, chest radiograph findings, and the chest CT scan findings, including imaging of the adrenal glands. Additional imaging studies (eg, brain CT scan, bone scan, and positron emission tomography [PET] imaging) may be warranted on the basis of clinical findings to delineate file nature of distant disease. (2)

The clinical evaluation of the stage largely depends on the interpretation of the chest CT scan images. Chest CT imaging is the most widely available noninvasive staging modality for lung cancer and often guides the choice of additional, more invasive steps. However, CT imaging for the clinical staging of lung cancer has limitations with respect to its test performance characteristics. The accuracy of CT imaging for metastatic disease in the thorax has been evaluated extensively, from the standpoints both of selecting the patient and of using the nodal station as the unit of analysis. (3-7) Despite improvements hi chest CT scan resolution over the last decade, its accuracy for the evaluation of nodal disease in the mediastinum has not improved (sensitivity, 57%; specificity, 82%). (8) In addition, chest CT scanning lacks reliability for the prediction of direct tumor invasion into the chest wall or into mediastinal structures (T3 and T4 involvement), with a false-negative rate of 18% and a false-positive rate of 32%. (9)

Our prospective, cohort study focused on the identification of factors associated with a misclassification of the true stage by comparing the cTNM with the pTNM of veterans with early-stage NSCLC (cN0, cN1) who were evaluated in a large multidisciplinary thoracic oncology clinic (MTOC).

MATERIALS AND METHODS

Patients and Data Collection

Prospective clinical data were collected on all patients referred to the MTOC at the Durham Veterans Affairs Medical Center from September 1997 to April 2002. The study included only patients with early, clinical stage I/II (cN0, cN1) NSCLC who were deemed to be candidates for surgical resection. Patients with radiographic evidence of cN2 disease were excluded from this analysis. Also excluded were patients with early-stage, clinical NSCLC whose lung function or comorbidities precluded surgical resection or who declined surgery because of personal preferences.

Demographic information, imaging findings, diagnostic study findings, pulmonary function test results, clinical and pathologic staging, types of treatments, lead times to diagnosis and therapy, and follow-up information were prospectively collected. Information on the tumor size and its location was based on the interpretation of the last chest CT scan obtained immediately before undergoing thoracotomy. Survival information was obtained through the medical records system and/or from the National Death Registry at the end of the observation period. All data were purged of personal identifiers before analysis. The data collection was approved by the local institutional review board.

Determination of Clinical Stage and Treatment Triage

The MTOC physicians determined the cTNM using thorough patient history information, physical examination findings, baseline laboratory test results (ie, CBC count, chemistry studies, liver function tests, and lactate dehydrogenase and serum calcium measurements), chest radiograph studies, and the findings of chest CT scans that included imaging of the adrenal glands. A short-axis transverse lymph node diameter of > 1 cm on the chest CT scan was used as the threshold for defining abnormal lymph node enlargement. (10)

The MTOC team included a pulmoonologist, a radiation oncologist, a medical oncologist, and a radiologist. Treatment decisions were made in the presence of the thoracic surgeon after a review of the imaging studies and tissue diagnosis with the radiologist and the lung pathologist. All involved members reached a consensus regarding further diagnostic workups and therapeutic interventions. All patients underwent surgical exploration and systematic lymph node dissection, and had pTNM ascertained. Their fired stage was determined through a careful review of the operative findings and the pathology report. The cTNM was directly compared with the pTNM to determine possible disparities.

Statistical Analysis

Continuous variables were analyzed using the Student t test, and categoric variables were analyzed by [chi square] test. The Kamofsky performance status was split into the following two performance categories: normal to mildly reduced (80 to 100%); and moderately to severely reduced ([less than or equal to] 70%). (11) Logistic regression was used to define the odds ratio (OR) for predictors of upstaging or downstaging after surgical exploration and to control for confounders. The survival time was calculated as the time from the date of diagnosis to the date of death or to the end of the observation period. Survival was evaluated only as the proportion of patients who were alive at the end of the observation period and were compared using a [chi square] test because of the variable length of follow-up and the small number of patients in some stage groups.

RESULTS

One hundred trine male patients with early, clinical stage (I/II) NSCLC were identified. All patients had pTNM determined at the time of thoracotomy.