3.1 The evidence for Memokath‑051 comprises 16 studies that met the inclusion criteria outlined in the scope, all of which included only adults with benign or malignant ureteric obstruction. Six of these were comparative observational studies: 2 full papers, 3 conference abstracts and 1 clinical trial record and abstract. The other 10 studies were single-arm observational studies published as full texts. For full details of the clinical evidence see section 3 of the assessment report.
3.2 The external assessment centre (EAC) considered all of the comparative studies and 8 of the 10 single-arm studies to provide results which were relevant to the decision problem. The other 2 single-arm studies (Bach et al. 2013 and Boyvat et al. 2005) had less relevance to the decision problem because they did not have acceptable levels of internal or external validity, so they were not included in the analysis.
3.3 The quality of reporting across all the studies was generally poor. None of the studies provided adequate details on patient characteristics, stent insertion procedures, follow-up, statistical analyses and uncertainty around the results. Migration rates and clinical success were the most commonly reported outcomes but definitions of clinical success varied, so statistical pooling could not be done.
3.4 The 6 comparative studies compared Memokath‑051 with:
Allium stents (Bolton et al. 2015)
UVENTA stents (Kim et al. 2014)
double‑J stents (Granberg et al. 2010, Maan et al. 2010)
Resonance stents (Nam et al. 2015)
ileal ureteral replacement surgery (Akbarov et al. 2017).
Memokath‑051 stents had lower clinical success rates compared with Allium stents, UVENTA stents and ileal ureteral replacement surgery, but was comparable to double-J and Resonance stents. In a pooled analysis of migration rates, Memokath‑051 stents had the highest rates compared with Allium, double-J and UVENTA stents. However, the EAC advised that the results should be treated with caution, because the rates for the comparators are informed by fewer studies and patients than those for Memokath‑051.
3.5 None of the studies provided comparative data for stent removal and replacement. A pooled analysis of the Memokath‑051 treatment arms showed that 16.0% were removed and replaced, 17.7% migrated and 6.3% were encrusted. There was not enough evidence for a subgroup analysis of patients who could not have surgery or antegrade or retrograde insertion. Clinical success rates for Memokath‑051 ranged from 50% to 64% in people with benign ureteric obstruction and from 33% to 100% in people with malignant ureteric obstruction. In Kim et al. (2014), Memokath‑051 had similar clinical success rates to UVENTA in the benign population but statistically significantly inferior success rates to UVENTA in the malignant population.
3.6 The EAC concluded that the evidence for Memokath‑051 came mainly from small, poorly reported observational studies. It considered that only 2 comparative studies (Maan et al. 2010 and Kim et al. 2014) and 1 single-arm study (Zaman et al. 2011) had acceptable internal and external validity.
3.7 The company submitted 3 economic studies, 2 of which were excluded by the EAC. The EAC identified 2 other relevant studies (Gonzalez et al. 2011 and Zaman et al. 2012). It considered that although the studies were poorly reported and in a heterogeneous group of patients, the results indicated that Memokath‑051 is likely to be cost saving compared with double‑J stents.
3.8 The model submitted by the company was based on an unpublished analysis comparing Memokath‑051 with double‑J stents. The EAC replicated the company's model, making it fully executable, and modified it to improve its usefulness for decision-making. The EAC's main changes included:
extending the time horizon to 5 years
including reconstructive surgery and other metallic stents as comparators
adding the ability to report a break-even time point between Memokath‑051 and the comparators
including the risk of urinary tract infections
adding scenario analyses to model the risk of unplanned Memokath‑051 replacement in 4 scenarios
introducing deterministic sensitivity analyses.
The cost modelling includes planned replacement of double‑J stents after 6 months and no planned replacement of Memokath‑051 over the 5-year time horizon. The EAC model also included a monthly risk of unplanned replacement for both technologies based on clinical data. Further details are in section 4.2.2 and 4.5 of the assessment report.
3.9 The company's base-case results showed that compared with double‑J stents, using Memokath‑051 could save £4,156 per patient over 2.5 years. After the EAC revisions to the model, this saving fell to £1,619 per patient over 5 years.
3.10 Compared with reconstructive surgery, Memokath‑051 stents are only cost saving if no planned replacement is needed. The incremental cost per patient after 5 years ranged from £467 to −£1,009, depending on the extrapolation of unplanned replacements.
3.11 Planned stent replacement is the main cost driver for Memokath‑051 compared with other metallic stents:
Compared with UVENTA and Allium stents, Memokath‑051 stents are cost neutral in the worst case but potentially cost saving with more positive assumptions.
Compared with Resonance stents, Memokath‑051 stents are cost saving after 12 months.
The EAC advised that the comparisons with Allium and Resonance stents should be interpreted with caution, because they are based on assumptions instead of comparative clinical data.
3.12 The EAC's analysis suggested that Memokath‑051 stents may be a plausible cost-saving treatment option for ureteric obstruction in people who cannot have reconstructive surgery and who are expected to need a ureteral stent for at least 30 months.