Phenotype and Biology of Early Colorectal Cancers

Background
The increased detection of pT1 colorectal cancers (CRC) in the National Health Service Bowel Cancer Screening Programme (NHSBCSP) raises new concerns for clinicians. The aim of this study is to investigate the phenotypic features and biology of screened and symptomatic pT1 CRC and to assess current and new high risk features associated with lymph node metastasis (LNM). The second aim of this study is to investigate the inter-observer variation of reporting screened pT1 CRC between pathologists.

Methods
Symptomatic and screened pT1 CRC were identified from two databases (Northern and Yorkshire Cancer Registry and Information Services [NYCRIS] and NHSBCSP database). Phenotypic features of the pT1 CRC were evaluated and compared from both cohorts.
The second part of the study investigated the inter-observer variability in the qualitative and quantitative assessments of screened pT1 CRC. Participating pathologists were asked to perform quantitative and qualitative assessments on 41 screened pT1 CRC. The level of agreement was determined using Fleiss Kappa statistics and intraclass correlation coefficient testing.

Results
Symptomatic CRC with LNM had a significantly wider area of invasion (p=0.001), a greater area of submucosal invasion (p < 0.001) and a higher proportion of tumour stroma (p = 0.005) compared to CRC without LNM. Symptomatic pT1 CRC were also significantly bigger in size than screened pT1 CRC.
The inter-observer variation study showed that quantitative factors had better levels of agreements than qualitative factors.

Conclusion
This study has shown that screened pT1 CRC are quantitatively smaller to their symptomatic counterparts suggesting that the NHS BCSP detects earlier pT1 CRC. This study also showed that novel quantitative factors such as width of invasion, area of submucosal of invasion and PoTS could be used as valid parameters in determining the rate of LNM. Finally, this study highlights the need for better guidelines/definitions in the evaluation of screened pT1 CRC.