REBUTTAL FROM DRS. PATZ AND GOODMAN

	ARTICLE

TOP ARTICLE

While we are all eager to reduce lung cancer mortality, the fact remains that no screening technique has proved effective. Dr. Jett and others suggest that chest radiographs used in prior trials were too insensitive, and that CT can find lung cancer at a smaller, more curable size. To date no published study has found a correlation between the small size of the primary lesion and survival, and in fact lesions < 3 cm, regardless of size, appear to have the same outcome.

Dr. Jett further defines his perspective by referencing prevalence data from several recent CT screening trials. These studies indeed confirmed the well-established fact that CT detects more nodules than chest radiographs, but it remains to be seen whether this will reduce deaths from lung cancer. In fact, although these CT studies reported a relative increase in Stage IA disease, they did not demonstrate the required concomitant decrease in advanced stage disease to confirm a true stage shift, which is needed to produce a meaningful reduction in mortality.

Dr. Jett also discussed one of the major problems with CT screening, the innumerable indeterminate nodules it detects. He suggests that computer-aided diagnosis (CAD) may assist in this area, although realistically CAD will pick up even more nodules, and thus amplify the problem. Other techniques including volumetric nodule measurement are also suggested to deal with this dilemma, but this method has not been validated in a large number of patients and still does not obviate the necessity to evaluate every single nodule. It remains to be seen whether refined anatomic and morphologic nodule assessment will ever be effective.

Finally, Dr. Jett discusses the cost. He cites one abstract that predicts CT screening will be cost-effective, although we do not believe one can prove cost-effectiveness before the test has been shown to be effective. What is clear is that the cost will be enormous.

Dr. Jett raises a number of important issues, but at this time only prevalence data are available. We still suggest that the results from appropriate trials are needed before mass CT screening can be recommended.