The open-label phase III study (CONVERT) recruited adults with active MAC lung disease, defined as MAC-positive sputum or bronchoscopy cultures despite guideline based therapy (GBT) for at least six months. Patients taking Arikayce® were more likely to achieve negative sputum culture conversion than those assigned to placebo. The intervention group also showed increased negative side effects such as dysphonia and cough. The submitting company originally presented a cost utility analysis comparing ALIS (Arikayce®) plus GBT versus GBT alone (SMC2369), for the treatment of MAC lung disease (MAC-LD) adults who have failed to response to previous GBT (for at least 6 months) and who do not suffer from cystic fibrosis. GBT is a multi-drug regimen which included IV amikacin as per recommended guidelines, included in the model for 11.52% of the comparator arm GBT patients, derived from a UK survey. The economic analysis therefore considered ALIS (Arikayce®) as an additional treatment for all patients, that is, ALIS (Arikayce®) would be an additional treatment for 88.48% of the patients, and replace IV amikacin in 11.52% of the patients. However, SMC clinical experts advised that in the Scottish setting ALIS (Arikayce®) would be used as a replacement for IV amikacin. Therefore the treatment arm in the base case economic analysis did not align with likely Scottish practice. A revised analysis with ALIS (Arikayce®) only as a replacement for IV amikacin was provided and was the focus of the economic analysis in this resubmission. It should be noted that in order to estimate the outcomes for this comparison, the submitting company assumed that the benefits of ALIS (Arikayce®) versus IV amikacin were assumed to be the equal to those of ALIS (Arikayce®) versus GBT alone from their original modelling. A number of weaknesses were identified with the analysis; • The benefits of ALIS (Arikayce®) versus IV amikacin were assumed to be equal to those of ALIS (Arikayce®) versus GBT alone, with a concern that this may overestimate the incremental effect, and thus under-estimate the cost-effectiveness ratio. The company asserted that the levels of incremental effect over IV amikacin were likely to be similar to those modelled in the original base case versus GBT alone on the basis that the population in the CONVERT trial were heavily pre-treated and many (one-third) had been pre-treated with this regimen and it was found not to be effective. • The lifetime horizon is a potential limitation, possibly overestimating quality of life gains for patients in the amikacin arm who spend a much greater proportion of time in the cure state. The New Drugs Committee (NDC) at the time of the initial submission considered a 10-year time horizon may be more relevant reflecting on the patient population but it is noted that this shorter time horizon could curtail some benefits for patients who achieve a cure. • Recurrence to MAC positive from the MAC negative state was applied as a constant risk stratified by treatment arm until microbiological cure can be achieved. The risk is derived from the CONVERT post hoc analysis, which has a very small sample size and therefore is subject to considerable uncertainty: (n= 8/65 ALIS plus GBT) and (n=4/10 GBT alone). Additionally in terms of recurrence rates, the recurrence from microbiological cure to MAC positive was set as constant in the base case model and is associated with uncertainty. The company explained that recurrence is influenced by a host of factors, and as no data are available as to how it would change over time is was set as constant. • Utilities in the model are derived from a combined weighted average from CONVERT study monthly utilities for converters/non converters to obtain utility values for the MAC positive and MAC negative states, stratified by treatment arm. This approach does not account for the baseline difference in utilities between arms from the CONVERT trial, and potentially further biases the model by assuming that once MAC negative state is achieved, quality of life would still vary by treatment arm. • The utility value for the cure state in the model (value from CONVERT post hoc analysis - 0.92) is applied as a constant rate. This value has some uncertainty given the small sample size and is not reflective of published UK age adjusted population utility values. Given the cure state has longevity in the lifetime model, UK population age adjusted utility values are more appropriate. • There is some conflicting information regarding EQ-5D data, analysis and methodology used to derive utility weights used in the model. The clinical report from CONVERT notes no evidence of improvements in health-related quality of life during treatment with ALIS. However, the post hoc analysis of EQ-5D data from CONVERT used in the model shows utility benefits for converters with ALIS compared to converters with GBT. The company suggest that a utility regression model would not appropriately capture the dynamic nature of the health state populations in CONVERT and also highlight the psychometric validation of the EQ-5D-3L data which demonstrated that the EQ5D-3L data had poor responsiveness to clinically meaningful changes in patients with MAC-LD. • Adverse events experienced by ALIS (Arikayce®) patients may be underplayed in the model, which only includes 4 AEs, which impact on cost parameters only. The submission notes that including additional disutility for early treatment related side-effects and adverse events could incur double counting. However, it is unclear that the combined weighted utilities derived from the CONVERT post-hoc analysis are a true reflection. The submitting company did not include any scenario analyses to explore impact of adverse events on quality of life and when requested to do so reiterated that the treatment-specific utilities used in the model already account for disutilities due to AEs. The company however noted that the £32k ICER would not capture any quality of life advantages of ALIS (Arikayce®) over IV amikacin, such as reduced otoxicity. • SMC clinical experts advised that ALIS (Arikayce®) patients would require education and training on how to use the equipment and take medication - either at home or in an outpatient setting. The NHS related costs of this were not accounted for. It should be noted that while this could be an important service impact, including these costs would be unlikely to affect the ICERs given the high medicine acquisition cost which influences the ICER. It was also noted that there may be differences in costs and outcomes for a comparison to the use of nebulised amikacin injection and IV amikacin and this had not been explored in the presented results. The Committee considered the benefits of ALIS (Arikayce®) in the context of the SMC decision modifiers that can be applied when encountering high cost-effectiveness ratios and agreed that as ALIS (Arikayce®) is an orphan medicine, SMC can accept greater uncertainty in the economic case. After considering all the available evidence and the output from the PACE process, the Committee accepted ALIS (Arikayce®) for use in NHSScotland.