Conclusion

This second report of the Ovarian Cancer Audit Feasibility Pilot examines variation in the treatment of cases of ovarian, tubal and peritoneal cancer diagnosed in England between 2016 and 2018. The report describes differences in treatment by stage, age, morphology and comorbidity, then explores geographic variation in treatment with adjustment for these factors.

With regard to patient demographics and tumour characteristics (Appendix 1), findings from this report confirm what previous research and clinical practice have indicated, including that:

• Women with stage 4 disease or no stage recorded, and tumours classed as miscellaneous and unspecified, were much less likely to receive any treatment.
• Women with underlying medical conditions, identified by the Charlson comorbidity index, were less likely to receive surgery.
• Older women were more likely to have chemotherapy alone or not receive any chemotherapy or surgery.

Some of these differences can be explained by a range of underlying factors. For example, younger women may be more likely to have surgery without adjuvant chemotherapy due to a greater prevalence of certain tumour types among younger women. Likewise, tumours classed as miscellaneous and unspecified, or where no stage is recorded, can reflect disease that is too advanced for surgery, classification and staging.

Findings also point to worrying patterns of variation, especially in relation to age. Results reported in Appendix 3 indicate that the likelihood of receiving surgery may be far lower for older age cohorts than younger women, even after accounting for factors including stage and morphology. Some of this variation may be explained by factors including the burden of comorbidities not captured by the Charlson comorbidity index, poor performance status, and patient choice (such as opting for chemotherapy over surgery; Appendix 5). Research is needed to explore the reasons for diagnoses in older age groups having a lower probability of surgery.

In terms of geographic variation, differences beyond those that may be explained by random variation alone were present for all treatments investigated (any treatment versus no treatment; surgery versus no surgery; chemotherapy versus no chemotherapy; adjuvant versus neoadjuvant chemotherapy). This variation was particularly apparent for surgery, with cancers in six Cancer Alliances showing an above average probability and five showing a below average probability of surgery out of a total of 19 Cancer Alliances. These regional variations may be attributable to a variety of factors not accounted for in the maximally-adjusted models, including differences in access to primary care enabling early diagnosis and timely referral to secondary care. However, differences may also reflect real variation between gynaecological cancer centre multidisciplinary teams in the efficiency of diagnostic pathways or preparedness to perform radical surgery and/or administer chemotherapy to women with advanced disease.

While stressing differences in the time coverage, covariate adjustment and cohort definitions of the two sets of analyses, cross-referencing results from this study with the outputs of the first report of the Ovarian Cancer Audit Feasibility Pilot (Disease Profile in England: Incidence, mortality, stage and survival for ovary, fallopian tube and primary peritoneal carcinomas)⁴ indicates that Cancer Alliances that were less likely to undertake treatment generally had lower than average five year survival figures, and that this relationship may be particularly pronounced for surgery. Together, these findings present an opportunity for further work to better understand the reasons for variation in treatment between areas, the impact of this variation on patient health outcomes, and the steps that can be taken to address it.

With this analysis indicating geographic variation in the treatment of ovarian, tubal and peritoneal cancer after adjusting for important variables, future outputs of the Ovarian Cancer Audit Feasibility Pilot will start to answer some of the outstanding questions outlined above, focusing in greater detail on treatment variation and seeking to better understand the treatment pathway of those patients with the poorest survival.

The analysis presented in this report is based on data collected prior to the COVID-19 pandemic and provides a baseline measure of access to treatment across England. The pandemic has already had an immense impact on cancer diagnoses and access to treatment⁵ and, while it is too early to quantify its full impact on cancer diagnosis and treatment, it makes the need for a continuing and fully-funded ovarian cancer audit even more pressing given the worrying picture presented within this report.