The use of deep learning models to predict progression-free survival in patients with neuroendocrine tumors.

Aim: The RAISE project assessed whether deep learning could improve early progression-free survival (PFS) prediction in patients with neuroendocrine tumors. Patients & methods: Deep learning models extracted features from CT scans from patients in CLARINET (NCT00353496) (n = 138/204). A Cox model assessed PFS prediction when combining deep learning with the sum of longest diameter ratio (SLDr) and logarithmically transformed CgA concentration (logCgA), versus SLDr and logCgA alone. Results: Deep learning models extracted features other than lesion shape to predict PFS at week 72. No increase in performance was achieved with deep learning versus SLDr and logCgA models alone. Conclusion: Deep learning models extracted relevant features to predict PFS, but did not improve early prediction based on SLDr and logCgA.

Description of the use of multicriteria to support pricing and reimbursement decisions by European health technology assessment bodies.

BACKGROUND: Heterogeneity in drug access throughout Europe may be influenced by differences in drug-assessment strategies. The EUnetHTA's assessment core model (EUnetHTA-core) and the EVIDEM's multicriteria framework are reference methodologies in this context, the latter including a wider compromise between non-contextual and contextual criteria. Compliance of 37 European Health Technology Assessment bodies (HTAb) with EUnetHTA-core has been reported, but the use of EVIDEM by this HTAb is still unknown. METHODS: To describe the uptake and use of multicriteria approaches to evaluate drug value by European HTAb using EVIDEM as reference framework, a multicriteria framework was obtained based on EVIDEM model. The criteria used for drug appraisal by HTAb was extracted from the EUnetHTA report, and completed through search of websites, publications and HTAb reports. Use of EVIDEM assessment model in 37 European HTAb has been described semi-quantitatively and summarized using an alignment heatmap. RESULTS: Aligned, medium or misaligned profiles were seen for 24,3%, 51,4% and 24,3% of HTAb when matching to EVIDEM dimensions and criteria was considered. HTAb with explicit responsibilities in providing specific advice on reimbursement showed more aligned profiles on contextual and non-contextual dimensions. CONCLUSIONS: EUnetHTA's core model is limited in assessing medicines while EVIDEM's framework provides contextual dimension used by some HTAb in Europe that can be escalated to other agencies. Most of the 37 European HTAb have room to broaden their contextual assessment tools, especially when social and medical perception of need requires to be explicit to support payer's decision on reimbursement.

Reporting reimbursement price decisions for onco-hematology drugs in Spain.

Introduction: Even using well-established technology assessment processes, the basis of the decisions on drug price and reimbursement are sometimes perceived as poorly informed and sometimes may be seen as disconnected from value. The literature remains inconclusive about how Health Technology Assessment Bodies (HTAb) should report the determinants of their decisions. This study evaluates the relationship between oncology and hematology drug list prices and structured value parameters at the time of reimbursement decision in Spain. Methods: The study includes all new onco-hematological products (22), with a first indication authorized between January 2017 and December 2019 in Spain and pricing decisions published up until October 2022. For each product, 56 contextual and non-contextual indicators reflecting the structured multiple criteria decision analysis (MCDA) - Evidence-based Decision-Making (EVIDEM) framework were measured. The relationship between prices and the MCDA-EVIDEM framework was explored using univariate statistical analyses. Results: Higher prices were observed when the standard of care included for combinations, if there were references to long-lasting responses, for fixed-duration treatment compared to treatment until progression and treatment with lower frequencies of administration; lower prices were observed for oral administration compared to other routes of administration. Statistically significant associations were observed between prices and the median duration of treatment, the impact on patient autonomy, the ease of use of the drug, and the recommendations of experts. Discussion: The study suggests that indicators related to the type of standard of care, references to long-lasting responders, the convenience of the use of the drug, and the impact of treatment on patient autonomy, as well as contextual indicators such as the existence of previous clinical consensus, are factors in setting oncology drug prices in Spain. The implementation of MCDA-EVIDEM methodologies may be useful to capture the influence on pricing decisions of additional factors not included in legislation or consolidated assessment frameworks such as the European Network for Health Technology Assessment (EunetHTA) core model. It may be opportune to consider this in the upcoming revision of the Spanish regulation for health technology assessments and pricing and reimbursement procedures.

Relevance of outlier cases in case mix systems and evaluation of trimming methods.

OBJECTIVES: To determine the most appropriate outlier trimming method when the main source of information for case mix classification is length of stay (LOS) because cost information is unavailable. METHODS: Discharges (35,262) from two public hospitals were analysed. LOS and cost outliers were calculated using different trimming methods. The agreement between cost and LOS trimming was analysed. RESULTS: The trimming method using the geometric mean with two standard deviations (GM2) showed the highest level of agreement between cost and LOS and revealed the greatest proportion of extreme costs. Nearly 5% of cases were outliers, containing 16% of total LOS. This was the best approximation to 18% of extreme cost because when GM2 was applied to LOS, 88% of outlier cost was revealed. CONCLUSIONS: The methods were analysed because they are the most frequently used but the same methodology could be employed to compare other outlier determination methods. Outliers should be calculated because they ought to be valued differently from inlier cases.