Premium
Back to the future: routine morphological assessment of the tumour microenvironment is prognostic in stage II / III colon cancer in a large population‐based study
Author(s) -
Hynes Seán O,
Coleman Helen G,
Kelly Paul J,
Irwin Steven,
O'Neill Roisin F,
Gray Ronan T,
McGready Claire,
Dunne Philip D,
McQuaid Stephen,
James Jacqueline A,
SaltoTellez Manuel,
Loughrey Maurice B
Publication year - 2017
Publication title -
histopathology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.626
H-Index - 124
eISSN - 1365-2559
pISSN - 0309-0167
DOI - 10.1111/his.13181
Subject(s) - medicine , hazard ratio , colorectal cancer , oncology , stromal cell , population , confidence interval , stage (stratigraphy) , cancer , cohort , pathology , biology , paleontology , environmental health
Aims Both morphological and molecular approaches have highlighted the biological and prognostic importance of the tumour microenvironment in colorectal cancer ( CRC ). Despite this, microscopic assessment of the tumour microenvironment has not been adopted into routine practice. The study aim was to identify those tumour microenvironmental features that are most likely to provide prognostic information and be feasible to use in routine pathology reporting practice. Methods and results On the basis of existing evidence, we selected specific morphological features relating to peritumoral inflammatory and stromal responses, agreed criteria for scoring, and assessed these in representative haematoxylin and eosin (H&E)‐stained whole tumour sections from a population‐based cohort of 445 stage II / III colon cancer cases. Moderate/severe peritumoral diffuse lymphoid inflammation and Crohn's disease‐like reaction were associated with significantly reduced risks of CRC ‐specific death [adjusted hazard ratio ( HR ) 0.48, 95% confidence interval ( CI ) 0.31–0.76, and HR 0.60, 95% CI 0.42–0.84, respectively]. The presence of >50% tumour stromal percentage, as assessed by global evaluation of tumour area, was associated with a significantly increased risk of CRC ‐specific death ( HR 1.60 95% CI 1.06–2.41). A composite ‘fibroinflammatory score’ (0–3), combining dichotomized scores of these three features, showed a highly significant association with survival outcomes. Those with a score of ≥2 had an almost 2.5‐fold increased risk of CRC ‐specific death ( HR 2.44, 95% CI 1.56–3.81) as compared with those scoring zero. These associations were stronger in microsatellite instability ( MSI )‐high tumours, potentially identifying a subset of MSI ‐high colon cancers that lack characteristic morphological features and have an associated worse prognosis. Conclusions In summary, reporting on H&E staining of selected microscopic features of the tumour microenvironment, independently or in combination, offers valuable prognostic information in stage II / III colon cancer, and may allow morphological correlation with developing molecular classifications of prognostic and predictive relevance.