The Get Data Out programme routinely publishes cancer statistics produced by NHS D (previously PHE) in a consistent table, called the Get Data Out (GDO) table. This table collects patients into groups with common characteristics, and then publishes information such as incidence, treatment rates, survival and Routes to Diagnosis for these groups.
This document sets out the definitions of the cohort and groups for the Get Data Out tables for the 2023 release of data on oesophageal and stomach cancers between 2013 and 2020.
The cohort of oesophageal and stomach cancers used for Get Data Out is all cancers coded in ICD 10 to C15 or C16.
This cohort includes malignant gastrointestinal stromal tumours (GISTs) of the stomach and oesophagus. The cohort does not include GISTs coded with non-malignant behaviour codes in ICD 10/O-2: these account for almost 75% of stomach GISTs in England between 2013 and 2018 (please see sarcoma for a full GIST breakdown).
Cancers were classified as oesophagus or stomach by site (ICD 10) and morphology (ICD-O-3.1), as documented in table 1 below and Appendix 1. Due to cancer registry coding improvements for tumours of the oesophagogastric junction (OGJ), the tumour type definition differs by year of diagnosis. This should be noted when undertaking time-series analysis on the data. The increase in the number of oesophagus tumours and decrease in stomach tumours in 2018 compared to 2017 (around 800 tumours) is partly due to this change in definition. All OGJ tumours have moved from the stomach to the oesophagus.
For tumours diagnosed between 2013 and 2017, oesophagus is defined as C15 with the additional inclusion of C16.0 squamous cell carcinomas (SCC) (ICD-O-3.1 8050-8086). Stomach is defined as C16 excluding C16.0 SCC (ICD-O-3.1 8050-8086). For tumours diagnosed from 2018 onwards, oesophagus is defined as C15 and C16.7, and stomach is defined as C16 excluding C16.7.
Between 2013 and 2017, grouping C16.0 SCC tumours with the oesophagus is more appropriate than grouping them in the cardia and oesophagogastric junction (OGJ) stomach group. This is because ICD 10 C16.0 does not distinguish between the cardia (part of the stomach) and the OGJ (where the oesophagus and stomach join). It is likely that tumours of C16.0 that have a SCC morphology are truly oesophageal tumours as this morphology is prominent in the oesophagus and rare in the stomach. Over 95% of malignancies in the stomach are adenocarcinomas (ADC) and squamous cell carcinomas are very rare in the stomach. For diagnoses from 01.01.2018, cancers of the OGJ are coded to C16.7. These tumours are staged and treated more like tumours of the oesophagus, so are moved into the oesophagus group. For diagnoses from 01.01.2018, cancers coded to C16.0 are true cardia cancers (even if they are rare SCC’s), so are left in the stomach.
Oesophageal and stomach cancers were further classified into subsites by (ICD 10) and morphology (ICD-O-3.1), as documented in table 2 below and Appendix 1. Where stomach tumours are partitioned by tumour type 2, stomach GISTs will not be reported on as a separate group. Please see the GDO sarcoma group for complete data on GISTs.
The classification of the lower third of the oesophagus changes in 2018 as cardia and OGJ tumours now have separate ICD10 codes. Until 2018, the lower third of the oesophagus is defined as C15.2, C15.5 and C16.0 SCC (ICD-O-3.1 8050-8086). From 2018, the lower third of the oesophagus is defined as only C15.2 and C15.5, with C16.0 SCC moving into the cardia group of the stomach. From 2018, tumours of C16.0 (now cardia only) are coded to tumour type 2 cardia, and tumours of C16.7 (now OGJ only) are coded to tumour type 2 oesophagogastric junction.
As the definitions of some groups change in 2018, 2016-2018 and 2017-2019 combined years data containing these different definitions should not be used. The groups that have different definitions in 2018 are:
To obtain the best approximate matching definition for time-series analysis pre and post-2018, tumours of the OGJ with an ‘ADC’ or ‘Other’ morphology should be moved into the stomach cardia group, and tumours of the OGJ with a ‘SCC’ morphology should be moved into the oesophagus lower third group. This method does not mean the definitions pre and post-2018 match exactly, and care should still be taken when undertaking time-series analysis on the data. There will be some C16.0 SCC tumours left in the stomach cardia group that would have been moved into the oesophagus lower third group pre-2018, but this number will be low.
Oesophageal cancers were classified by morphology (ICD-O-3.1) into adenocarcinomas, squamous cell carcinomas or other. Stomach cancers were classified by morphology (ICD-O-3.1) into adenocarcinomas or other. Please see table 3 below and appendix 2 for details.
Where groups are large enough a breakdown by stage at diagnosis was used. Where possible individual stages have been used:
For smaller groups with a stage split, the following stage breakdown was used:
The registration of 2019 tumours were being completed during the COVID-19 pandemic. This led to reduced access to the usual data sources, and despite the registry’s best efforts a noticeable decrease in data quality in some fields. This is most commonly seen in an increase in ‘stage unknown’ tumours, and a corresponding decrease in other stage groups. This should be noted when undertaking time-series analysis on the data.
The vast majority of cases are staged in either the UICC 7 or UICC 8 system, where the recommended staging system changed from UICC 7 to UICC 8 between 2017 and 2018 diagnoses. A very small percentage of cancers (less than 1%, apart from UICC 6 which accounted for 1.259% of cases in 2013), were staged in other systems, as reported in the known limitations page. The following table shows the percentage of cases staged in ENETS, UICC7 and UICC8 over time.
The final partition is by age of the patient at diagnosis. Only the remaining groups which contain a large enough patient count were split by age. Stomach groups large enough, were partitioned by age splits of:
This is to show the association with frailty in patients of 80+. Smaller groups were split into age groups of:
or for the cardia groups:
For oesophagus, groups were split by a general age split of:
Where groups with an ADC morphology were large enough to split by three age groups rather than two, a younger age split to reflect the association with Barrett’s oesophagus was used:
All classifications were created with Mr William Allum from The Royal Marsden, Alan Moss from Action Against Heartburn, and Dr Daniela Tataru and Dr Brian Rous at NCRAS.