| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Letters to the Editor |
New York Presbyterian Hospital- Weill Cornell Medical Center New York, New York
In Response: In response to Drs. Silvestri and Bach, it was clearly stated that all eight patients with clinical stage I lung cancer who had no treatment died of lung cancer in our response to the letters to the editor in the New England Journal of Medicine (1). The details of the procedures for obtaining the cause and timing of deaths have already been given (2). In his editorial, Dr. Aisner noted the importance of following those who had no treatment for their lung cancer. This continues as a critical aspect of the I-ELCAP as it shows the natural course of screen-detected lung cancer in the absence of treatment. Surely we need to continue to accumulate additional patients without treatment or with delayed treatment as they are identified. Beyond this, we address potential "overdiagnosis" in several ways (2) by (a) requiring growth of nodules <15 mm before biopsy, (b) having an expert pathology panel review all pathology specimens, and (c) following those who have no treatment or who delay their treatment. Ultimately, the benefit of screening comes from the opportunity to provide for early treatment and, in the absence of performing a trial that randomly assigns patients with screen-diagnosed lung cancer to immediate versus delayed treatment, our quasi-experimental design provides this information.
The I-ELCAP study provides a new paradigm for performing screening research by studying the diagnostic and prognostic issues separately (3). This approach has advantages over the traditional screening randomized control trial. Far from being a "case series" as suggested by Dr. Silvestri, each screenee is enrolled according to a common protocol (4) and is followed for all rounds of screening to determine whether they have lung cancer, either diagnosed by the screening or by prompting of symptoms between the screenings. Once a person is diagnosed with cancer, they are followed to learn the curability of each relevant subtype of cancer. Curability is estimated by finding the asymptote on the Kaplan-Meier survival curve (2). It is well known that once screening stops, the deaths will start occurring at the same rate as in the absence of screening, and these deaths do not contribute to the knowledge of the benefit of screening.
Dr. Bach is concerned that we referred to an article we wrote that explains the time delay before a mortality reduction can be seen in a breast cancer–screening program (5). We used the Malmo breast cancer randomized screening trial to show that the benefit of early treatment provided by screening was only seen some 8 years after the start of screening, and only as long as screening continued. This article was recognized as the turning point in the mammography controversy in 2002 where failure to understand this delay in the benefit nearly led to a change in public policy (6, 7). This article is particularly on point with regard to the Bach et al. article in JAMA (8) where the focus is on mortality reduction in the early years of follow-up where no reduction is to be expected.
References
Related Articles
Correction
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
