Article Text

Download PDFPDF
PTH-170 Evaluating the relationship between clinicopathological variables and mortality in patients with esophagogastric adenocarcinomas
  1. MA Butt1,
  2. R Haidry2,
  3. V Sehgal1,
  4. D Graham1,
  5. M Gandy3,
  6. M Arez1,
  7. E Bloom1,
  8. S Khan4,
  9. I Puccio4,
  10. D Oukrif5,
  11. M Banks2,
  12. R Hamoudi4,
  13. M Rodriguez-Justo5,
  14. L Lovat1,
  15. M Novelli5
  1. 1Research Department of Tissue and Energy, University College London
  2. 2Gastroenterology and Endoscopy, University College Hospital
  3. 3Pathology, The Doctors Laboratory
  4. 4Integrated Molecular and Cell Science Laboratory
  5. 5Pathology, University College London, London, UK


Introduction Gastric (GC) and oesophageal adenocarcinoma’s (OA) are a major global health problem. This study analysed the relationship between demographic, clinical and pathological variables on and all-cause mortality in a cohort of GC and OA patients.

Method Data was captured from 227 patients for age, sex, highest histological grade in tissue, tumour type, tumour location, TNM stage, HER2 status, R0 resection, nodal status, presence of linitis plastica, inflammation, atrophic gastritis (for GC) or Barrett’s epithelium (BE) (for EC). All patients’ received standard treatment through a multidisciplinary team and tissue specimens were examined by expert pathologists. Mortality was the primary outcome and was measured from date of diagnosis to date of death from any cause to calculate median survival (MS) of assessed variables. Kaplan-Meir curves were then compared between related variables (log-rank test, Mantel-Cox) and (log-rank test for trend).

Results Using Mantel-Cox analysis, MS (months=m) was significantly worse in GC vs EA (12.5 m, n = 126; 23.8 m; n = 51; p = 0.027); diffuse vs intestinal type tumours (17.4 m, n = 69; 27.9 m, n = 97; p = 0.008); poorly vs moderately differentiated tumours (20.78 m, n = 51; 32.68 m, n = 125; p = 0.08); metastatic vs localised cancers (6.5 m, n = 28; 26.4 m, n = 64; p= <0.0001); and absence of background inflammation vs its presence (17.29 m; n = 74; 28.8 m; n = 59; p = 0.047). For patients with EA, the presence of BE was associated with a worse MS when compared with cases without BE (15.15 m, n = 15; 41.12 m, n = 17; p = 0.0018). HER2 status, gender, linitis, atrophic gastritis (in GC only) or R0 resection rates did not significantly affect MS. Using the log-rank test for trend for graded variables, MS was worse for older age, ≥80 yrs (3.98 m; n = 23), 60–79 yrs (25.08; n = 104) or <60 yrs (27.12; n = 45)(p= <0.0001); progressive tumour stage, T1 (43.98 m, n = 30), T3 (23.93 m, n = 44) and T4 (17.75 m, n = 48) (p = 0.007); progressive nodal stage, N0 (80.7, n = 39), N1 (23.8 m n = 62), N2 (20.8 m, n = 11) or N3 (17.6 m; n = 8) (p < 0.0001) but not by Siewert stage (I, II or III) for junctional tumours.

Conclusion Survival was worse in tumours of gastric origin, diffuse grade, poor differentiation, higher TNM stage, and for OA, in those arising from Barrett’s epithelium. Older age was a worse prognostic indicator. This data helps understand the prognostic implications of GC and OA histology when counselling patients.

Disclosure of interest None Declared.

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.