Article Text


Pathology free papers
OC-105 Impact of AJCC 7th edition TNM staging on a historical oesophago-gastric cancer resection dataset
  1. A M Reece-Smith1,
  2. M P John1,
  3. Z Chaudry2,
  4. S L Parsons1
  1. 1Department of Oesophago-gastric surgery, Nottingham University NHS trust, Nottingham, UK
  2. 2Department of Histopathology, Nottingham University NHS trust, Nottingham, UK


Introduction The revised AJCC1 TNM staging systems for oesophageal and gastric cancer were published in 2010 and have been adopted in many units since that time. However, such changes in staging system can cause problems in comparing recent datasets to historical data as stage matching will no longer be consistent. We aimed to review historical data to determine the impact of updated scoring systems on our dataset, firstly to determine the proportion of patients changing stage groups and secondly to determine if survival was different between new and old systems.

Methods A database of gastro-oesophageal resections has historically accumulated patients with operative stage recorded using 6th edition TNM classification. With the help of a specialist pathologist these cases were reassessed to determine the revised TNM according to the 7th edition. The survival of matched stage disease was compared using Kaplan–Meier analysis with log-rank test for statistical significance.

Results In a cohort of 358 patients 50 patients (14.0%) changed stage. Twenty-four to a lower stage and 26 to higher stages as detailed in the Abstract OC-105 table 1 below. Stage 2 and 3 contained sufficient patients for survival analysis. Median survival was not reached for stage 2 in TNM6 and was 37.9 months in TNM7 (p=0.651). In stage 3 the survival was 16.7 months in TNM6 and 17.3 months using TNM7 (p=0.786). Survival was not significantly different between editions.

Conclusion Fourteen percent of patients change stage using updated criteria. However, the impact of these changes on the median survival of patients in specified stage groups is small. Re-evaluating historical patient data will not greatly advise clinicians or patients regarding their prognosis but this data does help in the comparison of historical publications to current data.

Abstract OC-105 Table 1

Competing interests None declared.

Reference 1. Ann Surg Oncol 2010;17:1721–4.

Statistics from

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.