Lung cancer is estimated to result in close to a million deaths from cancer annually worldwide. Majority of patients are diagnosed late and the 2-year survival is dismal despite improving chemotherapeutic options. However, the 5-year survival is close to 70% if it can be diagnosed early. So there is a pressing need to have effective screening tools for lung cancer to detect it early.

Given the magnitude of possible mortality reduction, the question remains is that can screening be done? Several studies have investigated this in randomized controlled fashion but to date, none of the studies utilizing chest X-ray, sputum cytology or CT scans have demonstrated an overall disease-specific mortality reduction.

The Mayo Lung Project that screened male smokers between 1971 and 1983 that used chest X ray (CXR) and sputum cytology failed to demonstrate a reduction in lung cancer mortality with extended follow-up despite an improvement in survival in the patients in the screened arm. Several reasons and observations to account for this are advanced from these early studies:

• A skin lump that has been present for some time but is growing slowly or becomes ulcerated. This is a very common presentation of a basal cell carcinoma.
• Some of the small primary lesions screened had metastatic disease already at the time of screening
• An earlier diagnosis of cancer did not lead to a delay in death (lead-time bias)
• Some of the cancers detected on screening had little clinical impact on the patient even if undetected.

Even with low dose CT screening the number of cancer diagnosed as a percentage of positive screens is low, ranging from 2-4%. Nevertheless it is important to emphasize that the rates of cancer diagnosis from positive screens for smokers are consistently higher than for non-smokers.

Conventional wisdom dictates that current screening methodologies have not resulted in reduction of cancer mortality. Another problem with screening is that it results in discovery of abnormalities that warrant invasive and potentially risky interventions even though a significant proportion of them are benign.

More definitive results can be expected in the next 5 years from data maturing from current trials, such as the National Lung Screening Trial (NLST) that has just completed accrual of 50,000 current or former smokers in the US and compares the role of CT scan versus CXR for screening.

 

Darren Lim
Consultant
Medical Oncology
National Cancer Centre, Singapore