NICE commissioned an external assessment centre (EAC) to review the evidence submitted by the company. This section summarises that review. Full details of all the evidence are in the project documents on the NICE website.
3.1 The evidence assessed by the EAC included 19 studies reported across 20 full text publications. Of the included studies, 5 were comparative (3 randomised controlled trials [RCTs] and 2 observational studies) and 14 were single-arm observational studies. The comparative evidence included a total of 595 people with intermediate-risk or high-risk non-muscle-invasive bladder cancer (NMIBC), of whom 247 had treatment with mitomycin C (MMC) using Synergo. Nineteen abstracts identified were not included in the evidence review. For full details of the clinical evidence, see section 3 of the assessment report in the supporting documentation on the NICE website.
3.2 Two RCTs compared treatment with MMC using Synergo with Bacillus Calmette-Guerin (BCG) therapy in intermediate-risk and high-risk NMIBC patients (Arends et al. 2016 and the HYMN trial [Tan et al. 2019]). Arends et al. assessed treatment with Synergo first line, including in people with intermediate-risk cancer who would not normally be offered BCG first line in the NHS. The HYMN trial was a UK-based RCT that included people with NMIBC for which BCG treatment had failed. Colombo et al. (2003 and 2011) compared MMC using Synergo with MMC alone in people with primary or recurrent intermediate-risk and high-risk NMIBC, with 10‑year follow-up data.
One RCT showed significantly better disease-free survival with MMC using Synergo compared with MMC alone
3.3 In the trial with a 10‑year follow up (Colombo et al. 2011) disease-free survival was significantly better with MMC using Synergo than with MMC alone (p<0.004). But there was no significant difference in overall survival (p=0.558). The 2 RCTs comparing MMC using Synergo with BCG showed no difference in recurrence-free survival (Arends et al. 2016) or disease-free survival (the HYMN trial). In people with non carcinoma in-situ (CIS) recurrence (papillary tumours only), the HYMN trial showed a non-significant difference in disease-free survival with Synergo compared with BCG (53% compared with 24%; p=0.11).
3.4 All 3 trials stopped early for various reasons: Colombo et al. because of significantly better efficacy with MMC using Synergo, the HYMN trial because of higher than expected CIS recurrence rate in the Synergo arm, and Arends et al. because of slow recruitment.
All trials used a low-dose adjuvant regimen so some people with CIS may have had treatment that was not effective enough
3.5 All trials only offered an adjuvant regimen (two 30‑minute cycles of 20 mg MMC) so 68% of people in the HYMN trial and 22% in Arends et al. with CIS may have not had effective enough treatment. In practice they would have had a higher ablative dose (two 30‑minute cycles of 40 mg MMC). Colombo et al. (2003, 2011) included only 1 patient with CIS so most people in this trial are likely to have had treatment with an appropriate regimen. Whether the ablative regimen using Synergo is more effective than other treatment options in people with CIS cannot be determined from the evidence currently available.
3.6 The HYMN trial was considered to most accurately reflect Synergo use in the NHS. This is because it was a UK-based RCT which included people with NMIBC for whom BCG treatment had failed. The study also included mostly people with high-risk cancer (87%). The EAC noted that the HYMN trial had several issues, which limits the quality and certainty of the results. Not all people in the comparator arm had treatment with BCG. The comparator was BCG or standard care, so some people had treatment with MMC alone or MMC-EMDA (electromotive drug administration of MMC). More people in the Synergo arm had concurrent papillary and CIS tumours, which have a higher risk of recurrence and progression. And the trial did not report on the type of BCG failure before enrolment, although the numbers who had fewer than or more than 6 instillations were reported.
One retrospective comparative study reports post-cystectomy outcomes in people who have had previous treatment with Synergo
3.7 Sri et al. (2020) was a retrospective cohort study that included people who had radical cystectomy for high-risk NMIBC. It compared outcomes between people who had a:
primary cystectomy, or cystectomy immediately after BCG failure (102 people) and
cystectomy after treatment with MMC using Synergo, after BCG failure (36 people).
The study reported no significant difference in the time to recurrence or mortality (all-cause and cancer-specific) between the 2 groups. Results suggested that delaying a cystectomy to have second-line treatment with Synergo did not worsen oncological outcomes compared with having the cystectomy straight away and no treatment with Synergo. But relatively few people in the study had treatment with Synergo, and the EAC considered it to have a high risk of selection bias.
No comparative study looked at high-risk NMIBC alone and no distinction can be made between the results of the different risk groups
3.8 All studies included people with intermediate-risk and high-risk NMIBC and in most cases the results were not reported separately. The extent to which results can be generalised to a high-risk group only is therefore uncertain.
3.9 Fourteen non-comparative studies reported on treatment with Synergo in people with intermediate-risk and high-risk NMIBC. Overall, the non-comparative studies were considered to be of low to medium methodological quality. This was because of, for example, retrospective analyses, small patient numbers, lack of comparators, limited outcomes reported, unclear reporting of risk classifications and, in some cases, uncertainty about whether there was patient overlap between studies. Only 2 were considered prospective studies (Erturhan 2015 and Kiss 2015) and 2 included UK centres (Sooriakumaran 2016; Van Valenburg 2018). There was a high level of heterogeneity in patient characteristics, treatment schedule and follow-up time. Recurrence rates (reported in 13 studies) varied depending on whether an ablative or adjuvant regimen was used, whether patients had had previous BCG treatments, and whether patients had concomitant CIS.
Adverse events appear to be mild to moderate and transient, with few patients stopping treatment because of side effects
3.10 Outcomes related to safety, tolerability and adverse events of Synergo therapy were reported in 18 studies and overall were reported to be mild to moderate and transient with few patients stopping treatment because of side effects. The most common adverse events during treatment included pain and spasms. After treatment, the most common adverse events were painful or difficult urination (or both), urination at night, and increased urinary frequency. For full details of the adverse events, see section 6 of the assessment report in the supporting documentation on the NICE website.
3.11 The company submitted a de novo cost analysis, which compared MMC using Synergo with MMC alone in people with intermediate-risk and high-risk NMIBC, for whom BCG is either unavailable or unsuitable. It used a Markov model comprising 4 health states: remission, recurrence (treated with radical cystectomy in all cases), post-cystectomy, and death. The model had a 1-year cycle length and a lifetime time horizon. The population age was 64 years. BCG was not included in the model as a comparator or as part of the clinical pathway. This was because the company considered it inappropriate to use the available comparative evidence between MMC using Synergo and BCG. Overall, the company's model showed that, compared with MMC alone, treatment with MMC using Synergo was associated with a cost saving of £4,466 per patient over a lifetime time horizon. For full details of the cost evidence, see section 4 of the EAC's assessment report in the supporting documentation on the NICE website.
The company model results are robust but limited by their relevance to the positioning in the clinical pathway
3.12 The company's one-way sensitivity analysis showed that cost saving estimates were most sensitive to changes in the cost of Synergo, the risk of recurrence and the cost of stoma management. None of the variations in parameters made treatment with Synergo cost incurring over a lifetime time horizon.
3.13 The EAC updated the company model and treatment with Synergo was found to be cost saving by £3,549 per patient over a lifetime horizon. The key drivers of the model were the cost of Synergo, the risk of recurrence, stoma management and the cost of cystectomy. The EAC noted that in the NHS it is unlikely that the decision would be between using MMC with Synergo and MMC alone.
Additional modelling by the EAC shows that MMC using Synergo is likely to be cost incurring compared with second-line BCG
3.14 The EAC did another additional analysis that better reflected current NHS use of the technology by only including people with recurrence of NMIBC after BCG therapy. Treatment with MMC using Synergo was modelled as an alternative to further BCG therapy in people with intermediate-risk or high-risk NMIBC whose disease recurred after intravesical therapy with BCG. The EAC amended the base case model using data from the subgroup analysis of patients without CIS in the HYMN trial (n=33), who were considered to have had treatment with the appropriate dose (adjuvant regimen). Except for costing for BCG therapy instead of MMC alone, all other costs were unchanged. The EAC removed the adverse events costs for MMC using Synergo, because the cost of adverse events was assumed to be similar for MMC using Synergo and BCG therapy. Mortality parameters remained unchanged.
3.15 Compared with BCG as a second-line treatment for patients with no CIS, treatment with MMC using Synergo was associated with an increased cost per patient of £9,858 over a lifetime horizon. Key drivers of the model were treatment costs, annual recurrence rates and starting age. None of the 20% variations in parameters made treatment with MMC using Synergo cost saving over a lifetime horizon.
Treatment with MMC using Synergo is associated with fewer cystectomies and an increase in life years gained and QALYs
3.16 All models resulted in fewer radical cystectomies and an increase in total life years and quality-adjusted life years (QALYs). In the model comparing treatment with MMC using Synergo with MMC alone, the incremental reduction in cystectomies was 0.22 per person and the increases in total life years and QALYs per person were 2.15 and 2.35, respectively. In the model that compared MMC using Synergo with second-line BCG, these changes were smaller: a 0.02 per person reduction in cystectomies, and a 0.8 and 0.79 increase in the total life years and QALYs per person, respectively.
Exploratory analysis suggests that MMC using Synergo is likely to be cost incurring compared with cystectomy in NMIBC that does not respond to BCG
3.17 The EAC did an exploratory analysis on the cost impact of treating NMIBC that did not respond to BCG with MMC using Synergo compared with radical cystectomy. It was based on the additional modelling for Synergo compared with second-line BCG therapy. It showed that treatment using Synergo was cost incurring by £12,180 per patient compared with cystectomy but was associated with a gain in life years and avoided cystectomy in 4% of people, modelled over a lifetime horizon. For full details, see the addendum to the EAC's assessment report in the supporting documentation on the NICE website.