Utility of the PET Scan in Lung Cancer
A Case Study

By Mohammed Moinuddin, M.D., F.A.C.P..

A 62 year old white female was evaluated for chest mass. The CT scan of the chest performed elsewhere, showed a right upper lobe mass and mediastinal lymph nodes. The patient underwent a bronchoscopy, which identified an endobronchial lesion in the right upper lobe. Bronchial washings, bronchoalveolar lavage and biopsy were performed. All three specimens were negative for neoplasm. The patient was referred for Positron Emission Tomography (PET). This (Fig.1) showed a sizable area of intense activity in the right upper lobe and a smaller area in the right paratracheal region. These measured 11.5 and 3.5 on standardized uptake value (SUV) scale. A small area of uptake was seen in the right parotid region and another lesion in the upper sacrum, probably S1. These findings were consistent with neoplasm.

With several biopsies being negative, the decision to be aggressive in the diagnostic workup and push for a more invasive approach were prompted by a strongly positive PET scan. The PET scan in this patient showed additional abnormalities which were suspicious for metastatic disease. Therefore, a mediastinoscopy and biopsy were performed. This showed adenocarcinoma.

Comment:
PET scan has recently been approved by the Committee on Medicare and Medicaid Services (CMS), previously known as HCFA, in the diagnosis and staging of lung cancer. There are approximately 150,000 cases of single pulmonary module (SPN) discovered in USA every year. Of these, 30-40% are malignant. Radiologic criteria for malignancy include larger size, irregular margins, cavitation, thick walls, absence of calcification, etc. However, these factors give an index of probability for malignancy and are not definitive. Biopsy is the only absolute way of diagnosing these nodules.

Since the PET approval by CMS in August 2001 in the workup of SPN, this diagnostic modality has been used extensively. Though not perfect, PET carries a significantly higher accuracy for diagnosing SPN compared with CT, chest-x-ray and MRI. The metabolic activity in SPN can be determined by glucose (FDG) uptake in the lesion expressed on the PET scan as a standardized uptake value (SUV). A value of >2.5 is regarded as consistent with malignancy and <2.5 as inflammatory. However, in a minority of patients, higher SUV may be seen in non-malignant diseases such as granulomatous infections, sarcoidosis, bacterial infections, etc. and low grade tumors may have SUV less than 2.5. Overall, the sensitivity, specificity and accuracy of PET in diagnosing lung cancer are 96%, 88% and 94% respectively for SPN>1cm. Smaller than 1 cm. lung cancers may be false negative on PET scans.

Some of the articles in the literature seem to indicate that PET scan findings may give more exact information than the biopsy at times. None of the procedures in medicine carry 100% accuracy and therefore exceptions do occur. For example, bronchial washings carry 75% accuracy for adenocarcinoma and >90% for squamous cell carcinoma. Bronchoalveolar lavage has a better yield for bronchoalveolar carcinoma because it lines the epithelium but has a lower sensitivity for other histological types. Bronchoscopy and biopsy have even higher accuracy (>95%).

PET scan covers not only the region of interest but also other areas which may harbor metastatic disease. Therefore, it may give additional information that may be crucial to make the diagnosis.

Hence, patients in whom the biopsy is negative (because of sampling problem etc.) and the PET scan is strongly positive for neoplasm, the diagnosis should be aggressively pursued with a repeat biopsy as in this case. This patient had three negative biopsies - bronchial washings, BAL and a bronchoscopic biopsy. The strongly positive PET pushed for a mediastinoscopic biopsy which clinched the diagnosis.