Hepatic arterial embolisation for patients diagnosed with neuroendocrine neoplasms: experience from a European Neuroendocrine Tumour Society centre of excellence

Background: Hepatic arterial embolisation (HAE) is a recommended treatment option for patients with neuro-endocrine neoplasms (NENs) and liver metastasis. We aimed to describe outcomes after HAE for NENs at a single United Kingdom centre and investigate the influence of tumour characteristics.

Methods: Retrospective analysis of sequential patients with NENs undergoing HAE at The Christie NHS Foundation Trust between 2004 and 2018. Primary end points were symptomatic and radiological response; secondary end-points included overall-survival (OS), progression-free survival (PFS) and adverse events.

Results: Embolisation was performed in 47 patients: median follow up was 25 months; mean age 63 years (range 34–82); primary site gastrointestinal (64%), pancreas (17%), lung (11%), or unknown (8%); functional tumour in 70% (33 patients, of whom 21 had diarrhoea, 28 flushing, and 5 wheeze); 98% well differentiated tumour; grade 1/2/3 in 49%/45%/4% (maximum proliferation index 29%); liver only disease in 9 patients (20%), and 1/≥2 other metastatic sites in 47%/34%. Therapies received prior to HAE were somatostatin analogue (all patients), chemotherapy (16), targeted therapies (12), radionuclide therapy (5), and debulking surgery (4). At least 2 prior therapies were received by 27 patients, and ≥3 in 18 patients. Indications for HAE were: symptom control in 21% of patients, disease progression (38%), or both (40%). Bilobar HAE was performed in 64% of patients (unilobar 35%). Where documented, symptomatic response was experienced by 76% of patients. Partial radiological response was achieved in 15%, and disease stabilisation at 6 months in a further 57%. Median PFS was 9.8 months (95% confidence interval (CI) 6.5–14.8). Patients with functional tumours had significantly longer PFS; 14.5 months (95% CI 8.0–31.5) vs 6.1 months (95% CI 4.4–6.9), multivariable hazard ratio 0.35, 95% CI 0.15–0.81, P=0.02. No differences in PFS were observed between patients with grade 1 and grade 2 tumours, or between different primary sites. Median OS was 30.0 months (95% CI 12.2–63.3). Adverse events were experienced by 29 patients; most commonly pain (13 patients, 45%).

Conclusion: HAE provides symptomatic response and disease control in the majority of patients with NENs, and those with functional tumours may experience differential benefit.