Memokath 051 Ureter stent for ureteric obstruction

The EAG identified 16 studies (6 comparative studies and 10 single-arm studies) including 6 publications of 5 studies submitted by the company at guidance review. The comparative studies all had a retrospective design, with a small sample size of up to 27 people (Akbarov et al. 2017) in each treatment arm. The single-arm studies were observational case series, with sample sizes ranging from 4 people (Boyvat et al. 2005) to 73 people (Papatsoris and Buchholz 2010). For full details of the clinical evidence, see section 3 of the original assessment report in the supporting documentation (Newcastle upon Tyne Hospitals and York Health Economics Consortium External Assessment Centre, 2017).

Memokath 051 Ureter stent's clinical success rates ranged from 43% (Kim et al. 2014) to 100% (Granberg et al. 2010, Zaman et al. 2011). Results of the comparative studies suggested that the clinical success rate of Memokath 051 was comparable to double‑J stents (100% success rate in both arms; Granberg et al. 2010) and Resonance stents (82% and 86% respectively; Nam et al. 2015) but was lower than Allium stents (81% compared with 100%; Bolton et al. 2015), and UVENTA (43% compared with 82%; Kim et al. 2014). The definition of clinical success and how it was measured were not consistently reported in the studies.

Evidence from 2 comparative studies showed that Memokath 051 remained in place longer than double‑J stents (17 months compared with 4 months; NCT00166361, 2014) and UVENTA (14 months compared with 12 months; Kim et al. 2014). The average length of time in place using Memokath 051 was 11 months (Papatsoris and Buchholz 2010).

The evidence from the comparative studies suggested that stent migrations were higher with Memokath 051 than double‑J stents (11% against 0%; Maan et al. 2010) and UVENTA (43% against 6%; Kim et al. 2014). Encrustation rates were also higher with Memokath 051 than with double‑J stents (29% against 0%; NCT00166361, 2014) and Allium (19% against 0%; Bolton et al. 2015). The other most common adverse events included urinary tract infection (Akbarov et al. 2017, Klarskov et al. 2005, Papatsoris and Buchholz 2010, Zaman et al. 2011) and blockage or obstruction (Akbarov et al. 2017; Kim et al. 2014; Klarskov et al. 2005; NCT00166361 2014; Zaman et al. 2011).

The EAG did a pooled analysis, the results of which suggested that the stent migration rate for Memokath 051 was 17.7% (13 studies) compared with 5.9% using UVENTA (1 study) and 0% using double‑J stents (2 studies) and Allium (1 study).

The evidence was reviewed in 2021 and long-term retrospective data suggested higher complication rates with Memokath 051 than were previously reported, so NICE decided to update the guidance.

For the guidance update, the EAG considered 7 new studies including 2 abstracts:

• 1 systematic review and meta-analysis (Khoo et al. 2018)

• 2 retrospective non-randomised single-centre comparative studies (Choi et al. 2019, Khoo et al. 2021)

• 4 retrospective single-arm single-centre studies (Bier et al. 2017, Diaz Romero et al. 2018, Elbaroni et al. 2020, Forster et al. 2021).
  
  For full details of the clinical evidence, see section 4 of the assessment report update in the supporting documentation (King's Technology Evaluation Centre [KiTEC], 2022).

Evidence from a comparative retrospective study of people with chronic benign ureteral strictures reported that primary success rates (maintaining patency without additional procedures) were 28.6% for Memokath 051 and 12.0% for UVENTA during the observation period (Choi et al. 2019). The overall success rates (success defined as maintaining patency after further salvage procedures) were 57.1% for Memokath 051 and 40.0% for UVENTA. The differences between the 2 stents were not statistically significant. This result is consistent with the original guidance, which showed Memokath 051 had similar clinical success rates to UVENTA in the population with benign ureteral strictures.

The new evidence suggested that the median duration of actual functional stent follow up (censored by stent failure, death or end of study) was 5.5 months for Memokath 051, 11.4 months for Allium and 11.7 months for Resonance. This is shorter than the 2 studies included in the previous guidance, which reported 14 months and 17 months of indwelling time (Kim et al. 2014; NCT00166361 2014). Results of single-arm studies reported a median indwelling time of 11.8 months (range 1 week to 70.8 months; Bier et al. 2017) and a median stent lifespan of 14.5 months and 13.4 months in people with malignant ureter obstruction and benign ureter obstruction respectively.

Forster et al. (2021) reported long-term outcomes using Memokath 051 in an NHS centre. Only 25 out of 100 people included did not have any complications during a 5‑year follow up. The common complications were stent migration (36%) and failed ipsilateral upper tract drainage (27%), which included blockage (14%), encrustation (11%) and lost renal function (2%). The study included a subgroup analysis of benign and malignant ureteric obstruction, and the overall complication rate was significantly higher in people with a benign obstruction (85.4%) than those with a malignant obstruction (62.7%). Stent migration was the most common complication in people with a benign obstruction (53.7%) and failed renal drainage was common in people with a malignant obstruction (30.5%). A comparative study with a 5.5‑month follow-up period for Memokath 051 suggested that it had higher migration, obstruction and infection rates than Resonance, regardless of whether it was a benign or malignant obstruction (Khoo et al. 2021). Duration of follow-up was censored by stent failure, patient death or end-of-study period. The study did not have long-term comparative data on the complication rates of different stents.

The original guidance included 3 economic studies, all of which compared the cost associated with using Memokath 051 with double‑J stents (Aintree University Hospital 2012; Gonzalez et al. 2011; Zaman et al. 2012). The results indicated that Memokath 051 was likely to be cost saving compared with double‑J stents, although the studies were poorly reported and included a heterogeneous group of people with varying types of obstruction and life expectancy.

The company and EAG search did not identify any new published economic studies since the original guidance. The EAG considered the evidence from 4 new clinical studies including 2 UK studies (Forster et al. 2021, Khoo et al. 2021), 1 German study (Bier et al. 2017) and 1 Korean study (Choi et al. 2019) to be relevant for updating the risk of unplanned stent replacement in the economic model.

The company developed a simple cost model in the original guidance. The EAG thought that the company's cost model captured the key aspects of treatment, but that the way it dealt with certain structural issues was too simplistic, such as only including double‑J stents as a comparator. The EAG adapted the company's model and made the following main changes to the original guidance:

• extending the time horizon from 2.5 years 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.
  
  Scenario analyses were also added to model the risk of unplanned Memokath 051 replacement. More details of the cost model are in sections 4.2.2 and 4.2.3 of the original assessment report in the supporting documentation (Newcastle upon Tyne Hospitals and York Health Economics Consortium External Assessment Centre, 2017).

The EAG considered that the original model structure and assumptions remained valid. It updated the model's parameters because of new evidence available:

• An increase in the monthly risk of unplanned stent replacement for Memokath 051 from 1.40% to 1.80% compared with double‑J, Allium, Resonance and reconstructive surgery. A decrease in the monthly risk of unplanned stent replacement for Memokath 051 from 4.40% to 3.57% compared with UVENTA for chronic benign ureteral strictures.

• The monthly risk of unplanned stent replacement remained unchanged for double‑J stents (0%). But the monthly risks of unplanned stent replacement were updated from:

• 0.49% to 5.54% for Allium

• 1.40% to 1.78% for Resonance

• 0.49% to 4.99% for UVENTA.
  
  The model assumed that if someone needed an unplanned stent replacement it automatically delayed the planned replacement by the length of time in situ (thus removing any double counting). For full details of the clinical parameters, see economic model parameters in section 9.2 of the assessment report update in the supporting documentation (KiTEC 2022).

The cost of Memokath 051 and related consumables did not change. The EAG updated the cost of double‑J stents, UVENTA and Resonance according to the NHS Supply Chain reported unit prices. The cost of Allium was the same as the original guidance. The EAG also updated other costs such as staffing and follow-up visits using the most up-to-date data sources available. For full details of the cost sources, see economic model parameters in section 9.2 of the assessment report update in the supporting documentation (KiTEC, 2022). In the original guidance, a passport balloon dilator was needed when inserting Memokath 051, and the assumption was that a dilator was also needed when inserting UVENTA and Resonance stents. At consultation, the EAG ran a scenario analysis and found that removing the cost of a balloon dilator for Resonance had a minimal effect on overall costings and did not change the cost saving conclusion for Memokath 051.

The EAG's updated base-case results showed that using Memokath 051 is cost saving per person over 5 years by:

• £1,926 compared with double‑J stents

• £6,260 compared with Resonance

• £8,813 compared with UVENTA

• £9,365 compared with Allium.
  
  But using Memokath 051 incurred an additional £1,321 per person over 5 years compared with reconstructive surgery. Memokath 051 was also cost incurring compared with reconstructive surgery in the original guidance (£467) but the size of the additional cost has increased. Base-case results were modelled over a 5‑year period, reflecting the indwelling duration for Memokath 051 after which planned replacement is needed. Risks of unplanned replacement were also applied for the full-time horizon of the model.

The EAG's scenario analyses showed that, compared with UVENTA, Allium and Resonance, Memokath 051 was cost saving in all scenarios. Compared with double‑J stents, Memokath 051 was cost saving in all scenarios, except when the replacement cost of double‑J stents dropped to £717. Memokath 051 would be cost neutral if the replacement cost of a double‑J stent was £1,008. In addition, Memokath 051 would be cost incurring compared with double‑J stents if the monthly unplanned replacement risk was 2.81% over a 2‑year time horizon. Scenario analyses were done to investigate the impact of unplanned replacement of Memokath 051 at various timepoints in the stent lifespan, but it was found to still be cost saving compared with UVENTA, Allium and Resonance. For full details of the scenario analyses, see economic model parameters in section 9.3 and 9.4 of the assessment report update and appendix D of the addendum to the assessment report update in the supporting documentation (KiTEC, 2022).

Compared with reconstructive surgery, Memokath 051 was cost incurring in all the scenarios, except when the reconstructive surgery cost was £12,656 and the Memokath 051 monthly follow-up cost was £21 (1 visit per year). The threshold at which Memokath 051 could be cost neutral was if the cost of reconstructive surgery was £9,287. Memokath 051 would be cost saving if its risk of replacement in the first 2 years was 1.8%, and it was not replaced in the remaining 3 years of a 5‑year time horizon, or it had a 1.8% monthly unplanned replacement risk over a 2‑year time horizon.