Background:

Lutathera® has been reimbursed in the UK since 2018 for treatment of unresectable or metastatic, progressive, well-differentiated, somatostatin positive gastroenteropancreatic neuroendocrine tumours (GEP-NETs) in adults. This analysis aimed to explore healthcare resource use and costs in England for progressive NET patients treated with Lutathera® and matched patients treated with somatostatin analogues (SSAs), chemotherapy, or targeted therapies (everolimus or sunitinib).

Methods:

We analysed the Hospital Episodes Statistics (HES) database for two cohorts of NET patients (using ICD-10 coding) from 2016-2018. The Lutathera® cohort was defined as NET patients with a procedure code for imaging or SSA treatment and subsequent code for radionuclide therapy. A progressive NET cohort was defined as patients with increased frequency of SSA therapy, switch to chemotherapy or targeted therapies, or metastatic cancer code subsequent to initiating SSA. HRG codes were used for costing.

Cohorts were matched on propensity scores with sex, age at disease progression and Charlson Comorbidity Index as parameters.

Results:

The matched cohorts consisted of 199 patients. Patients were well matched on all characteristics with a mean Pscore difference of 0.00397.

From 2016-2018, Lutathera® demonstrated lower overall costs (£1.52M Lutathera vs £2.37M non-Lutathera, p=0.0003), non-elective spells and costs (289 vs 611 days and £850K vs £1.7M, p<0.0001 respectively) and A&E costs (£42K vs £62K, p=0.0013). Lutathera® demonstrated numerical benefit in average length of stay for elective and non-elective stays (14.2 days vs 23.3 days). For the Lutathera® cohort number of elective spells (195 vs 80, p<0.0001) and associated costs (£374K vs £242K, p<0.0001) were higher due to in-hospital dosing.

Conclusions: These exploratory analyses indicate significantly lower overall costs and non-elective hospital stays for progressive NET patients treated with Lutathera®. Given limitations of HES, further analyses will incorporate additional datasets (CPRD and/or NCRAS) and allow estimation of survival and insight into treatment pathways.