Authors' objectives: The National Institute for Health and Care Excellence is piloting a new approach to evaluating health technologies, which takes into consideration the full treatment pathway for a condition. This report describes the first pilot topic for the pathways process, which evaluated systemic treatments for advanced renal cell carcinoma. This pilot aimed to develop a decision model representing the treatment pathway that will be used to evaluate new technologies for advanced renal cell carcinoma. The pilot also evaluated a new treatment for renal cell carcinoma: cabozantinib (Cabometyx®; Ipsen, Slough, UK) plus nivolumab (Opdivo®; Bristol Myers Squibb, Princeton, NJ, USA). The National Institute for Health and Care Excellence (NICE) is piloting a new approach to evaluating health technologies, which takes into consideration the full treatment pathway for a condition. The ‘pathways’ process aims to increase the efficiency of reimbursement decisions in the NHS. The approach is based on developing comprehensive and adaptable core models for specific disease areas to which new treatments can be added and compared over time. The approach can also be used to evaluate optimum treatment sequences. This report describes the first pilot topic for the pathways process, which evaluated systemic treatments for advanced renal cell carcinoma (aRCC), which includes locally advanced and metastatic renal cell carcinoma (RCC). The project aimed to develop an evidence base to inform the development of a decision model representing the treatment pathway. The model is available open source without restriction, allowing reuse for future appraisals while maintaining confidentiality of proprietary data. As part of this phase of the pilot, the project evaluated a new treatment for aRCC, cabozantinib (Cabometyx®; Ipsen, Slough, UK) plus nivolumab (Opdivo®; Bristol Myers Squibb, Princeton, NJ, USA). The RCC originates in the tubules of the kidney and is the most common type of kidney cancer (80% of cases). Clear-cell RCC (ccRCC) is the most common subtype of RCC (around 75%), with the remainder comprising papillary, chromophobe and others. The disease is staged from 1 to 4, according to degree of spread. Stage 3 indicates locally advanced cancer (with regional lymph node involvement), and stage 4 indicates metastatic disease beyond the regional lymph nodes. Stage 3, locally aRCC that is unresectable with surgery may instead be treated with systemic treatment in the first line. Prognostic risk scores are available to predict survival in people with RCC, including the International Metastatic RCC Database Consortium (IMDC) risk model, which is used to inform systemic treatment options alongside prior treatment exposure. The objectives of this analysis were to develop an open source, core decision model for systemic treatments for aRCC and use this model to estimate the costs, effects and cost-effectiveness of: all-risk and favourable-risk populations: cabozantinib + nivolumab versus pazopanib (Votrient®; Novartis, Slough, UK) versus tivozanib (Fotivda®; Recordati, Hemel Hempstead, UK) versus sunitinib (Sutent®; Pfizer, Sandwich, UK) as a first-line systemic therapy in people with untreated aRCC intermediate-/poor-risk population: cabozantinib + nivolumab versus pazopanib versus tivozanib versus sunitinib versus cabozantinib versus nivolumab + ipilimumab (Yervoy®; Bristol-Myers Squibb, Princeton, NJ, USA) versus pembrolizumab (Keytruda®; Merck Sharp & Dohme, London, UK) + lenvatinib (Lenvima®; Eisai, Hatfield, UK) as a first-line systemic therapy in people with untreated aRCC. In addition, this analysis considered treatment options at second line and beyond, including axitinib (Inlyta®; Pfizer, Sandwich, UK), cabozantinib, lenvatinib + everolimus (Afinitor®; Novartis, Slough, UK), sunitinib, everolimus, pazopanib, nivolumab, and including tivozanib as an off-label treatment, in people with previously treated aRCC of any risk group. Consistent with NICE methods, the analysis did not consider the cost-effectiveness of treatments for RCC that are not routinely commissioned in the NHS. This includes avelumab plus axitinib, which is currently only available to people with aRCC through the Cancer Drugs Fund.

Authors' results and conclusions: Cabozantinib plus nivolumab was associated with better progression-free survival and overall survival than existing tyrosine kinase inhibitors as first-line treatment in the all-risk group. Using the list price of the evaluated interventions, the incremental cost-effectiveness ratio for cabozantinib plus nivolumab compared to the next non-dominated tyrosine kinase inhibitor monotherapy (pazopanib [Votrient®; Novartis, Slough, UK]) was £275,106 per quality-adjusted life-year in the all-risk population and was £379,222 in the favourable-risk population. Incremental cost-effectiveness ratios were relatively consistent across the base-case and scenario analyses. In the intermediate-/poor-risk population, the incremental cost-effectiveness ratio for pembrolizumab (Keytruda®; Merck Sharp & Dohme, London, UK) plus lenvatinib (Lenvima®; Eisai, Hatfield, UK) was £450,638 compared to cabozantinib; cabozantinib plus nivolumab and nivolumab plus ipilimumab were both dominated by cabozantinib and pazopanib monotherapy, respectively, in the base-case analysis. Quality-adjusted life-year gains were similar for cabozantinib plus nivolumab, pembrolizumab plus lenvatinib and nivolumab plus ipilimumab (Yervoy®; Bristol-Myers Squibb, Princeton, NJ, USA). Cabozantinib plus nivolumab was shown to be less effective and less expensive than pembrolizumab plus lenvatinib in most scenarios. This pilot demonstrated the feasibility of producing a reference model, which is open source and available to relevant stakeholders without restriction. This will improve consistency in the National Institute for Health and Care Excellence’s decision-making and allow for the evaluation of optimum treatment sequences for advanced renal cell carcinoma. During the technical engagement phase of the appraisal, some analyses were updated to include additional data or information. This summary reflects those findings. Systematic literature review of clinical effectiveness evidence One hundred and eighteen SLRs and meta-analyses and 30 RCTs (of which 6 were ongoing) were identified in the SLR. Of the 24 complete RCTs, earliest recruitment was in 2006 and the latest datacut was from December 2019. Trials included between 3 and 200 centres, with at least 14 including a UK centre. Sample sizes varied between 22 and 1110 participants. None of the studies were considered to be at low overall risk of bias: nine were considered at high risk and eight were considered to be unclear. Despite the moderate evidence base identified, most interventions were supported by only one trial and data quality was poor, resulting in a number of uncertainties. Notably, almost all evidence identified was based on a ccRCC population, which may be associated with improved treatment response relative to other histologies. There was a lack of high-quality data for time to treatment discontinuation (TTD) and time to next treatment (TTNT) alongside poor reporting of the subsequent therapies received by participants. It was not possible to explore the effect of adjuvant pembrolizumab on the effectiveness of first-line therapies. A total of 4 databases, 12 publications and 5 stakeholder submissions provided information on relevant RWE sources. Data extracted included treatment patterns, OS, PFS, TTNT and TTD. No data sources provided information on health-related quality of life or resource use/costs, although relative dose intensity was reported in one source which was used to calculate drug costs in scenario analysis. The availability of a decision model for systemic treatments for aRCC will improve consistency in NICE’s decision-making and allow for the evaluation of optimum treatment sequences for RCC. However, limitations in the evidence present challenges, and further modelling approaches may be needed to account for data scarcity associated with some treatments, such as complexities with dosing and titration, limitations of using PFS as a surrogate for OS and discrepancies between results generated from FP and PH NMAs. Optimal treatment sequences, the long-term effectiveness of combination treatments and the effectiveness of combination treatments in favourable-risk RCC also remain areas of uncertainty. This pilot has demonstrated the feasibility of (1) the production of a reference model capable of evaluating treatment sequences as well as the introduction of a new technology within the treatment pathway and (2) the incorporation of UK RWE into economic analysis. The pilot has also highlighted the importance of thoroughly describing areas of uncertainty which are often underdeveloped within the single technology appraisal process, such as the impact of subsequent treatments on outcomes and the importance of using data reflective of the UK population to describe baseline risk.

Authors' methods: A systematic literature review was conducted to identify evidence to inform effectiveness, safety and economic model development, including systematic literature reviews, randomised controlled trials, economic evaluations, utility studies and cost and resource use data. Real-world evidence was sought following the recommendations of the National Institute for Health and Care Excellence real-world evidence framework. Structured expert elicitation informed assumptions about overall survival and progression-free survival. Network meta-analyses were conducted to evaluate the clinical effectiveness of treatments. A de novo state transition model that was constructed with a partitioned survival analysis structure was also presented. The cost perspective of the model was that of the National Health Service and Personal Social Services; the time horizon was 40 years, costs and outcomes were discounted at 3.5% per annum and a 2022 price year was used. The model allowed sequences of up to four active lines of treatment. Most interventions were supported by only one trial and data quality was poor. Outcomes reported in clinical trials were generally more favourable than those reported in real-world evidence, suggesting that trials may overestimate treatment benefits. The External Assessment Group (EAG) conducted a systematic literature review (SLR) to identify relevant published evidence and real-world data sets. The methods used were consistent with NICE-preferred methods and best practice guidance for the conduct and reporting of SLRs. As no randomised controlled trials (RCTs) directly compared every comparator head to head, the EAG conducted a network meta-analysis (NMA) to indirectly compare treatments. Development of the decision model commenced with a review of published cost-effectiveness studies, structured expert elicitation to provide estimates of parameters for which no data existed, development of the model in R (The R Foundation for Statistical Computing, Vienna, Austria), populated with effectiveness, quality of life and resource use/cost data and reporting of incremental analyses. Systematic literature review of clinical effectiveness evidence Searches were conducted to identify previous SLRs and meta-analyses and RCTs published since the most recent relevant SLRs. Database searches were complemented by additional hand-searching of grey literature. Ongoing RCTs were identified by review of relevant trial registries. Relevant data were extracted from study reports into a bespoke database and assessed for quality using the Cochrane risk of bias tool (v2). The EAG received a submission of evidence from the manufacturer of cabozantinib, which was appraised and used to inform the broader project. This included SLRs and NMAs, which were reviewed and compared against the EAG’s own methods. New data from the company, and other stakeholders, were extracted and included in the EAG’s analyses. Relevant real-world evidence (RWE) was sought following the recommendations of the NICE RWE framework. Additional sources were identified from stakeholder submissions.

Project Status: Completed

URL for project: https://www.journalslibrary.nihr.ac.uk/programmes/hta/NIHR136008

Year Published: 2026

URL for published report: https://www.journalslibrary.nihr.ac.uk/hta/GJDL0327

URL for additional information: English

English language abstract: An English language summary is available

Publication Type: Full HTA

Country: England, United Kingdom

DOI: 10.3310/GJDL0327

Organisation Name: NIHR Health Technology Assessment programme

Contact Address: NIHR Journals Library, National Institute for Health and Care Research, Evaluation, Trials and Studies Coordinating Centre, Alpha House, University of Southampton Science Park, Southampton SO16 7NS, UK

Contact Name: journals.library@nihr.ac.uk

Contact Email: journals.library@nihr.ac.uk