Article Text


Clinical staging underestimates pathological stage in non-small cell lung cancer
  1. L Camporota
  1. Research Registrar, Guy’s & St Thomas’ NHS Foundation Trust, London, UK;

Statistics from

Clinical staging of lung cancer should establish a cTNM reliably predictive of the pathological stage. This study analysed 2994 lung cancers, operated on with curative intent in Spain between 1993 and 1997, with the aim of determining the agreement between clinical and pathological staging.

93% of patients were male with a 57.5 pack-year smoking history. In 29% of the cases lung cancer was an incidental radiological finding. 98% of the cancers were non-small cell (59% squamous) and 80% underwent complete resection (55% lobectomy/bilobectomy, 32% pneumonectomy). Of the 2994 patients initially included in the series, 2606 had a clinical staging, 2710 were classified using the pathological staging, and a clinicopathological comparison was performed in 2377 cases (79%). The clinicopathological agreement for stages IA–IIIB was 47% (Kappa’s index 0.248), similar to that found in other studies (35–55%). The highest agreement (75%) was achieved for stages IA and IB (Kappa’s index 0.56) and the lowest for stages IIIB (22%) and IIIA (8%). Clinical staging underestimated the pathological staging in 92% of stage IIA, 86% of IIIA, 74% of IIB, and 15% of IB tumours. Differences in staging protocols and in the characteristics of the population studied are likely to explain variability in the results between different studies. Future studies, using integrated positron emission tomography (PET)/CT scanning, are likely to result in a better agreement between clinical and pathological staging. This is important for treatment planning and the provision of accurate prognostic information.

View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.