Appendix 1 - Cohort definitions

Ovary, fallopian tube and primary peritoneal carcinomas (‘ovarian cancers’) were selected from the National Cancer Registration Dataset (NCRD)¹ if diagnosed in England between January 2015 and December 2019. Cases were identified according to the following ICD-10 and ICD-O-2 codes:

• C56 (malignant neoplasm of ovary)
• C57 (malignant neoplasm of other and unspecified female genital organs)
• C48 (malignant neoplasm of retroperitoneum and peritoneum) - excluding sarcomas: 8693, 8800, 8801, 8802, 8803, 8804, 8805, 8806, 8963, 8990, 8991, 9040, 9041, 9042, 9043, 9044, 8810, 8811-8921, 9120-9373, 9490, 9500, 9530-9582
• D39.1 (neoplasm of uncertain or unknown behaviour of ovary)

Only tumours diagnosed within female patients were included in the cohort.

This definition aligns with international ovarian cancer analyses and is designed to capture all ovarian cancers; although there may be variation in the coding of the originating site within this group, in practice their prognosis and treatment are similar. The inclusion of D39.1 means that the cohort includes all ‘borderline malignant’ ovarian cancer.

For the incidence analysis, the ovarian cancers were selected if they met the criteria as that outlined above.

The definition of ‘ovarian cancer’ used for mortality statistics in this report is C56-C57 in ICD-10. Data on mortality from ovarian cancer are derived from the ONS Mortality Extract which is in turn derived from causes of death recorded on death certificates, which do not include morphology information. Hence the more nuanced cohort descriptions (as used for incidence, stage and survival) which rely on tumour morphology information are not available for this statistic.

Borderline malignant ('borderline') ovarian tumours make up approximately 16% of the overall cohort of ovary, fallopian tube and primary peritoneal carcinomas. These tumours have historically been recorded as ovarian cancers, though their malignant potential is now understood to be lower than the rest of the group.

In this report, survival analysis is presented for cohorts including and excluding borderline tumours. Their exclusion is in line with international ovarian cancer analyses and avoids inflation of the survival estimates due to the better survival of the borderline group. Hence, it gives a clearer picture of the survival of women with non-borderline ovarian cancer.

Cases were identified per the codes under the 'Defining ‘ovarian cancer’' heading, but with the exclusion of all tumours at ICD-10 site code D39.1. This is consistent with National Statistics methodology.

This methodology also applied the following selection criteria:

Include

1. Patients aged between 15 to 99 years (inclusive)
2. The earliest relevant primary tumour documented during in the period of analysis
3. An invasive, primary, and malignant behaviour code (3) tumour

Exclude

1. Death certificate only (DCO) registrations
2. Missing or imputed gender, date of diagnosis, date of birth or age information

For the geographic variation in treatment analysis, the ovarian cancers were selected if they met the criteria as that outlined above.

Tumours identified via death certificate only were excluded from the analysis of geographic variation in treatment as associated patients would not have been referred for treatment.

Additionally, stage 1 tumours were excluded as trial evidence has not demonstrated a major survival benefit of chemotherapy for low-grade stage 1 tumours.