How Real-World Evidence in Melanoma HTA: Scientific Momentum and Methodological Maturity

Real-World Evidence (RWE) in Melanoma Health Technology Assessment (HTA): From Data to Decision-Making

Across the United Kingdom, France, Germany, and Canada, health technology assessment (HTA) agencies are entering a new phase of scientific maturity—defined not by the size of clinical trials but by the credibility and methodological rigour of real-world evidence (RWE). Nowhere is this more visible than in melanoma, a disease that has undergone one of the fastest therapeutic evolutions in modern oncology.

In just over a decade, immune checkpoint inhibitors (nivolumab, pembrolizumab, and ipilimumab) and targeted BRAF/MEK inhibitors (dabrafenib/trametinib and vemurafenib/cobimetinib) have transformed metastatic melanoma from a uniformly fatal cancer into one where long-term survival is achievable. Yet, the evidence that secures market authorisation is not always the same as the evidence that sustains reimbursement or informs long-term access.

Traditional randomised controlled trials (RCTs), while essential for demonstrating efficacy, often fail to capture how these therapies perform in real-world populations—elderly patients, those with comorbidities, or individuals with central nervous system (CNS) metastases. This disconnect between trial efficacy and real-world effectiveness has made RWE indispensable for modern HTA decision-making.

Today, NICE (UK), HAS (France), IQWiG and G-BA (Germany), and CADTH (Canada) are each developing distinct yet converging frameworks for the use of RWE in oncology appraisals. These frameworks seek to answer a common question: how can post-approval observational data, registries, and electronic health records provide robust, unbiased, and policy-relevant evidence that complements clinical trial results?

The emerging methodological consensus now emphasises transparency, bias adjustment, and lifecycle evidence generation—from early access programmes to post-reimbursement reassessments. This represents a decisive shift from viewing RWE as supportive evidence to recognising it as a scientific cornerstone of oncology (HTA). Melanoma has experienced an extraordinary therapeutic revolution over the past decade, particularly with the introduction of immune checkpoint inhibitors (e.g., nivolumab, pembrolizumab, ipilimumab) and BRAF/MEK inhibitors (e.g., dabrafenib/trametinib and vemurafenib/cobimetinib) in the United Kingdom.

United Kingdom (NICE)

The National Institute for Health and Care Excellence (NICE) has progressively embedded real-world evidence (RWE) in its oncology appraisal framework, particularly for advanced melanoma, where long-term survival and immune-mediated outcomes challenge traditional modelling assumptions. From 2011 to 2019, NICE used RWE in 11 of 14 melanoma technology appraisals (Llewellyn, 2020), covering both immune checkpoint inhibitors and targeted therapies.

A landmark example is TA366 (nivolumab + ipilimumab combination therapy), where post-marketing registry data were essential in validating survival projections beyond the 3-year follow-up reported in clinical trials such as CheckMate-067. Similarly, in TA558 (pembrolizumab for untreated advanced melanoma), real-world registry data—principally from the Systemic Anti-Cancer Therapy (SACT) dataset and the Public Health England National Cancer Registration and Analysis Service (NCRAS)—were used to refine extrapolation of the “long-tail” survival curve observed in immunotherapy responders. These data supported an updated cost-effectiveness model demonstrating sustained benefit and improved overall survival (OS) beyond the trial horizon.

NICE’s 2022 Real-World Evidence Framework formalises the agency’s expectations for RWE submissions. It explicitly endorses the integration of structured observational studies, disease registries, claims data, and external comparators, provided that methods for confounding adjustment (e.g., propensity score matching, inverse probability weighting, or regression adjustment) are robust and transparent. Within oncology, the advice is particularly relevant for estimating treatment durability, dose modifications, adverse event rates, and health-related quality of life (HRQoL) in real-world settings—parameters that directly influence incremental cost-effectiveness ratios (ICERs).

For melanoma, NICE now actively encourages early dialogue on RWE design through its Office for Market Access (OMA) and Scientific Advice Program, ensuring the alignment of data sources and analytical methods before submission. The integration of RWE from UK cancer registries and hospital electronic health records has thus evolved from being “supportive evidence” to a core component of value demonstration for high-cost immunotherapies. As a result, NICE melanoma appraisals increasingly rely on hybrid models combining clinical trial efficacy with real-world effectiveness data to support reimbursement and reassessment under the Cancer Drugs Fund (CDF).

France (HAS)

The Haute Autorité de Santé (HAS) occupies a central role in France’s health technology assessment and reimbursement system, combining clinical evaluation, medico-economic appraisal, and long-term real-world evidence (RWE) oversight. In 2021, HAS published a comprehensive Guide méthodologique pour les études en vie réelle, introducing standardised templates for protocol design, data collection, and analytical transparency in RWE submissions (HAS, 2021). This framework was developed to align post-market data generation with reimbursement re-evaluation cycles and to strengthen confidence in observational evidence for oncology and rare diseases.

Despite this progress, RWE acceptance in melanoma HTAs has historically been limited and conservative. From 2011 to 2019, merely 22% of melanoma technology appraisals included real-world evidence (Llewellyn, 2020). HAS requires a clear demonstration of methodological robustness, including minimising selection bias, using validated data sources, and relevance to the French healthcare context. Foreign or multi-country registries are generally disregarded unless augmented with local French data, indicating the agency’s focus on external validity within the French healthcare system (Système National des Données de Santé, SNDS).

A notable example is the CT14929 appraisal of vemurafenib + cobimetinib in BRAF-mutated metastatic melanoma. Here, retrospective analyses from the MelBase registry and hospital pharmacy records were used to estimate real-world treatment duration, dose intensity, and adverse-event frequency, which diverged meaningfully from pivotal trial data (coBRIM). These findings informed cost-effectiveness recalibration by revealing higher rates of early discontinuation due to toxicity and lower cumulative drug exposure—factors that directly reduced the incremental cost-effectiveness ratio (ICER) per quality-adjusted life-year (QALY).

Beyond clinical validation, HAS increasingly uses RWE to support post-listing price negotiations through the Comité Économique des Produits de Santé (CEPS). Real-world utilisation data—such as hospital stay duration, management of immune-related adverse events, and outpatient follow-up costs—feed into Budget Impact Analyses (BIAs) and inform price-volume agreements under the Accord Cadre CEPS-LEEM. For high-cost oncology drugs, including checkpoint inhibitors, this approach enables dynamic reimbursement adjustment as more mature survival and safety data emerge.

In 2023, HAS piloted new RWE-based reassessment mechanisms for melanoma immunotherapies (nivolumab, pembrolizumab) under its post-market data-collection agreements (suivi post-inscription), mandating registries such as MelBase and ONCORA to collect longitudinal survival, quality of life, and healthcare utilisation outcomes. These datasets are now being used to inform Service Médical Rendu (SMR) re-ratings and to model real-world Amélioration du Service Médical Rendu (ASMR) grades for renewals of reimbursement status.

Scientifically, France’s HTA model is moving toward integrated life-cycle evidence generation, where RWE validates clinical durability, safety, and resource use within a universal-coverage system. For melanoma, this evolution marks a shift from reactive post-approval surveillance to proactive economic stewardship—linking real-world performance directly to reimbursement sustainability.

Germany (IQWiG/G-BA)

Germany’s approach to real-world evidence (RWE) in oncology health technology assessment (HTA) has historically been conservative, prioritising internal validity and randomised controlled trials (RCTs) as the evidentiary gold standard for the AMNOG (§35a SGB V) early benefit assessment process. Under this framework, newly authorised drugs must demonstrate “added benefit” (Zusatznutzen) compared to an appropriate comparator therapy, with decisions led by the Institute for Quality and Efficiency in Health Care (IQWiG) and final resolutions issued by the Federal Joint Committee (G-BA).

While this system was designed around RCT evidence, the scientific and clinical realities of oncology—particularly in advanced melanoma—have exposed key limitations. For instance, crossover designs, small biomarker-defined populations, and rapidly evolving treatment sequences make randomisation difficult or even unethical. As a result, IQWiG and G-BA have gradually softened their stance, recognising that real-world data can fill critical evidence gaps when “dramatic effects” are observed or when RCTs are infeasible.

IQWiG formally acknowledges RWE in exceptional cases—typically when effect sizes are large (e.g., relative risk >10, p < 0.01) and biological plausibility is strong (Sammon et al., 2020). In oncology, this clause has been invoked for breakthrough therapies showing pronounced survival advantages or immunologic durability that defies standard comparators. However, such evidence must still meet the scientific criteria of internal validity, requiring robust confounding control (e.g., multivariable regression, instrumental variable analysis, or target-trial emulation) and transparent data provenance.

The 2020 IQWiG position paper marked a pivotal shift: it explicitly recognised the potential of routine data sources such as the Versorgungsdaten der Gesetzlichen Krankenversicherung (GKV), electronic health records, and disease-specific registries (e.g., the Deutsche Krebsregister). IQWiG now encourages their use in complementary analyses—to validate external comparators, inform treatment duration, or refine assumptions in cost-effectiveness models. However, the agency maintains that these data must be pre-specified in a study protocol reviewed under its General Methods (Version 6.1, 2022), ensuring transparency, reproducibility, and statistical adjustment for confounding.

For melanoma, this evolution has practical implications. RWE can now contribute to benefit reassessments when new long-term survival or toxicity data emerge, as seen in subsequent G-BA updates for nivolumab and pembrolizumab. These analyses integrate registry-derived survival curves and post-market pharmacovigilance data to reassess the magnitude of additional benefit and to support price renegotiations under §130b SGB V.

Economically, RWE is also beginning to inform cost-benefit analyses (Kosten-Nutzen-Bewertung) and Budget Impact Analyses (BIAs), particularly where real-world dosing intensity or adverse event management costs diverge from trial conditions. The Federal Ministry of Health (BMG) has recently commissioned studies to explore the integration of RWE into AMNOG follow-up evaluations, a clear signal that methodological rigidity is giving way to pragmatic scientific maturity.

In short, while Germany’s HTA landscape remains anchored in RCT evidence, it is scientifically evolving toward an RWE-enabled model—one where routine data, national cancer registries, and quantitative bias analysis serve as legitimate complements to trial evidence. This shift reflects an emerging consensus within the German HTA community: that validity and transparency, rather than study design alone, determine evidentiary credibility.

IQWiG’s 2020 guidance emphasises the role of routine data sources with a limited formal structure for submission.

Canada (CADTH)

The Canadian Agency for Drugs and Technologies in Health (CADTH) has become one of the most adaptive HTA bodies in integrating real-world evidence (RWE) into oncology appraisals. Between 2011 and 2021, RWE was cited in approximately 58% of CADTH melanoma submissions, a reflection of Canada’s decentralised but data-rich healthcare ecosystem.

In metastatic melanoma, RWE has been critical for evaluating immunotherapy sequencing, treatment duration, and cost-effectiveness within Canadian practice. Retrospective cohort analyses and chart reviews from provincial cancer agencies—particularly the Alberta Cancer Registry, BC Cancer, and Ontario Health (Cancer Care Ontario) datasets—have informed model assumptions around real-world overall survival (OS), progression-free survival (PFS), and dose intensity for agents such as nivolumab, pembrolizumab, dabrafenib, and trametinib.

In the CADTH reappraisal of pembrolizumab (pCODR, 2021), real-world registry data helped refine transition probabilities in the cost-utility model, reducing uncertainty in long-term survival extrapolations and improving the robustness of incremental cost-effectiveness ratios (ICERs). Similar RWE inputs have been used to evaluate the economic impact of combination immunotherapies, particularly where trial data were immature or subject to crossover.

Recognising this value, CADTH issued draft RWE guidance in 2022, encouraging manufacturers and provincial payers to pre-specify data sources, define confounder control strategies (e.g., propensity score weighting or Bayesian adjustment), and align analytical frameworks with HTA objectives. Crucially, CADTH distinguishes between real-world descriptive studies (for contextual evidence) and comparative effectiveness studies (for quantitative incorporation into cost models).

This structured methodology has advanced CADTH’s transition from static one-time appraisals to dynamic lifecycle evaluations that integrate RWE for reassessment and conditional funding. Melanoma has served as a pilot disease area where RWE informs both clinical guidance and budget impact analyses (BIAs)—capturing provincial variations in treatment uptake, resource utilisation, and real-world patient outcomes.

Addressing Confounding and Bias in Real-World Evidence

A persistent critique of RWE in oncology HTA is its susceptibility to unmeasured confounding—variables not captured in observational datasets that influence both treatment allocation and outcomes. This concern has historically limited the evidentiary weight of RWE in regulatory and reimbursement decisions.

However, methodological advances are now addressing these limitations. Quantitative bias analysis, E-values, and Bayesian sensitivity modelling—techniques long used in pharmacoepidemiology—are increasingly advocated for HTA submissions involving RWE. For instance, Sammon et al. (2020) demonstrated that in a real-world comparison of alectinib vs ceritinib in ALK-positive non-small-cell lung cancer, the E-value of 2.03 implied that an unmeasured confounder would have to double the risk of both treatment assignment and mortality to negate the observed benefit—an unlikely scenario, thus strengthening causal inference.

Applied to melanoma, these quantitative methods can reinforce the credibility of external control arm studies, registry-based comparisons, and treatment-pattern analyses, transforming RWE from supplementary context to scientifically defensible evidence. HTA bodies such as NICE, HAS, IQWiG, and CADTH increasingly expect these analyses to be incorporated into submission dossiers, reflecting the growing methodological maturity of RWE in oncology.

RWE Across the Technology Lifecycle

The most effective use of real-world evidence now extends beyond initial market access to the entire lifecycle of a technology (Oortwijn et al., 2019; Thokagevistk et al., 2024):

Early Access and Conditional Reimbursement – RWE underpins managed entry agreements such as the UK Cancer Drugs Fund (CDF) and France’s post-listing data collection (suivi post-inscription), providing empirical data for continuation or withdrawal decisions.
Re-assessment and Price Renegotiation – Mature survival and safety data from melanoma immunotherapies feed into periodic re-evaluations of added benefit or Service Médical Rendu (SMR), informing price-volume adjustments.
Health-System Readiness and Resource Planning – RWE quantifies hospitalisation rates, biomarker testing capacity, and workforce burden, guiding budget impact analyses (BIAs) and implementation economics.
Post-market Pharmacovigilance – Registry-based monitoring of immune-related adverse events supports risk-management plans and label updates, bridging pharmacovigilance and HTA evidence needs.

This lifecycle perspective—linking clinical outcomes, economic performance, and system sustainability—is transforming melanoma HTA into a continuous evidence ecosystem rather than a single evaluation event.

Scientific and Policy Outlook: From Acceptance to Expectation

The trajectory across the United Kingdom, France, Germany, and Canada shows a clear pattern: real-world evidence is no longer peripheral—it is integral to oncology HTA. Yet its acceptance depends on scientific transparency, local relevance, and quantitative rigour.

Agencies are converging on several best practices:

Co-develop protocol templates with HTA bodies to ensure data acceptability.
Use national registries and claims data to strengthen external validity.
Integrate adherence, utilities, and resource-use parameters into cost-effectiveness models.
Apply quantitative bias-adjustment techniques (E-values, array approaches, Bayesian methods).
Engage in early scientific advice to pre-align RWE methodologies with appraisal frameworks.

Melanoma has become the proving ground for this methodological evolution. With survival horizons now exceeding a decade and combination regimens reshaping the standard of care, RCT-only evidence is insufficient to capture the realities of modern oncology. The fusion of trial precision with real-world representativeness represents not just a methodological refinement but a paradigm shift in how health systems assess, price, and sustain innovation.

References:

Llewellyn, S. (2020). Real-world evidence in HTA appraisals of melanoma therapies. [LinkedIn Post]
Thokagevistk, K. et al. (2024). This study discusses how real-world evidence can support and enhance the findings of clinical trials in health technology assessment (HTA). J. Mark. Access Health Policy. [https://doi.org/10.3390/jmahp12020009]
Sammon, C. et al. (2020). Real-world evidence and unmeasured confounding in HTA. J. Comp. Eff. Res. [https://doi.org/10.2217/cer-2020-0112]
Oortwijn, W. et al. (2019). HTAi Global Policy Forum – RWE in HTA: From Theory to Action.
NICE (2022). Real-World Evidence Framework. [https://www.nice.org.uk]
HAS (2021). Guide méthodologique pour les études en vie réelle. [https://www.has-sante.fr]

1) What is real-world evidence (RWE), and how does it differ from randomised controlled trial (RCT) data in melanoma?

RWE is clinical evidence derived from routine care (registries, EHRs, pharmacy/claims, audit datasets). It captures effectiveness in unselected populations, treatment sequencing, dose intensity, adherence, and resource use. RCTs optimise internal validity (randomisation, blinding, protocolised follow-up) to estimate efficacy under ideal conditions but typically exclude older adults, ECOG 2–3, brain metastases, and significant comorbidity. In melanoma, immune checkpoint inhibitors (ICIs) produce long-tail survival and immune-mediated toxicity that are under-characterised in trials; RWE complements this by showing how outcomes translate in practice.

2) Why is RWE increasingly critical for melanoma HTA?

Melanoma therapies (PD-1, CTLA-4, BRAF/MEK) create heterogeneous and durable responses, with late immune-related adverse events (irAEs) and variable treatment duration. HTA bodies must project lifetime health gains and budget impact beyond trial follow-up. RWE improves:

External validity: outcomes in elderly, comorbid, CNS metastases.
Durability modelling: validates survival plateaus in ICIs.
Real-world utilisation: dose reductions, discontinuations, outpatient toxicity management.
Economics: informs ICER and BIA inputs (utilities, resource use, uptake curves).

3) How are RWE studies improving OS/PFS estimates for ICIs?

Through linked longitudinal registries (e.g., SACT/NCRAS, MelBase, provincial cancer datasets) that extend observation to 5–10 years. Methods include:

Mixture-cure and spline models to capture plateau effects.
Piecewise hazards and parametric survival (e.g., log-logistic, Gompertz) validated against RWD.
Conditional survival updates (S(t|landmark)).
These reduce extrapolation uncertainty and change cost-effectiveness when a cured fraction exists.

4) Which HTA agencies are most advanced in accepting RWE for oncology?

NICE (UK): dedicated RWE Framework (2022); routine use of SACT; structured advice pathways.
HAS (France): RWE methodological guide (2021); strong preference for French data (MelBase/SNDS); RWE in post-listing follow-up for price/scope reviews.
IQWiG/G-BA (Germany): RCT-centred; RWE accepted in exceptional “dramatic effect” cases and for follow-up reassessments; growing use of routine GKV data and registries.
CADTH (Canada): structured RWE use, clear confounding expectations, strong provincial registry linkages.

5) How does RWE influence ICER calculations in melanoma appraisals?

ICER = (C₁ – C₀) / (E₁ – E₀). RWE refines:

C (costs): real-world dosing, irAE management (steroids, endocrinology), imaging, admissions, day-case infusions.
E (effects): OS/PFS from survival models anchored to RWD; utilities from PROMs; disutility for irAEs.
Time on treatment (ToT) and stopping rules substantially affect drug cost per QALY. Real-world early discontinuation typically lowers costs and sometimes reduces benefit, changing the ICER directionally depending on durability.

6) What are the most important RWE data sources for melanoma HTA?

Cancer registries: NCRAS/SACT (UK), MelBase/ONCORA + SNDS (France), German Krebsregister, Canadian provincial registries (Ontario/BC/Alberta).
Hospital EHRs and oncology pharmacy systems (dose/cycle timestamps).
Claims/administrative: admissions, outpatient activity, imaging, endocrinology/dermatology follow-up.
PROMs repositories (EQ-5D-5L, EORTC QLQ-C30/QLQ-MC).
Toxicity databases (pharmacovigilance; ICSRs).
Quality hinges on linkage, data completeness, coding accuracy, and governance.

7) How do HTA bodies evaluate methodological quality in RWE?

They look for pre-registered protocols, a priori covariates, transparent causal diagrams, and robust confounding control:

Propensity scores (matching, IPTW, overlap weighting).
Doubly robust estimators (AIPW).
Target-trial emulation (explicit eligibility, time zero, censoring rules).
Instrumental variables were plausible.
They expect missing data handling (multiple imputation), sensitivity analyses (trimmed weights, alternative calipers), and diagnostics (SMD balance, positivity checks).

8) Why is quantitative bias analysis (QBA) important, and how is it done?

QBA makes unmeasured confounding explicit:

E-value: the minimum association strength an unmeasured confounder needs with treatment and outcome to explain away the effect.
Array approaches: vary plausible confounder prevalence/effect to show robustness ranges.
Bayesian QBA: priors on confounding parameters, posterior on treatment effect.
When E-values are large and arrays show stability, HTA confidence in causal inference increases, particularly for external comparator studies.

9) Can RWE replace RCTs for comparative effectiveness?

Generally no—but RWE can be determinative when RCTs are infeasible/unethical (small biomarker strata, rapid SOC evolution) or for reassessment. The optimal pathway is hybrid: RCT core efficacy + RWE for generalisability, durability, and economics.

10) How is real-world adherence measured and used in economic models?

Adherence and exposure are proxied by admin timestamps, delays, dose reductions, and stops. Models use empirical ToT curves, dose intensity distributions, and state costs linked to toxicity and monitoring. Per-patient per month (PPPM) drug cost is recalculated using real-world dose/schedule, altering ICERs and budget impact.

11) What role does RWE play in post-marketing reassessment and price renegotiation (France/Germany)?

France: RWE informs SMR/ASMR re-ratings; CEPS uses utilisation and outcomes to tune price-volume and capping. Post-listing suivi mandates registry submission (e.g., MelBase) with predefined endpoints.
Germany: G-BA reassessments integrate registry survival/toxicity for added benefit magnitude; outcomes feed §130b price negotiations. Routine data projects are expanding under BMG initiatives.

12) How does RWE contribute to Budget Impact Analysis (BIA) for ICIs?

RWE supplies real uptake curves, eligibility prevalence, line-of-therapy mix, duration on treatment, toxicity-driven service use, and substitution effects (e.g., targeted therapy to ICI shift). BIAs model payer-year costs: drug, administration, monitoring, irAE management, imaging, end-of-life care. Sensitivity scenarios include stepped adoption, biosimilar entry, and stopping-rule enforcement.

13) What statistical techniques are recommended for real-world comparative effectiveness?

Propensity score methods: matching (nearest neighbour/optimal), IPTW/SMRW, overlap weighting (reduces extreme weights).
G-methods: marginal structural models for time-varying confounding.
Target-trial emulation for immortal-time bias control.
Instrumental variables (e.g., centre-level preference) if valid.
Flexible survival: Royston-Parmar splines, mixture-cure models.
Competing risks (Fine–Gray) for melanoma-specific mortality vs other causes.

14) How do registries like MelBase improve HTA evidence in France?

MelBase provides granular staging, molecular markers (BRAF), ECOG, metastatic pattern (incl. CNS), dosing, toxicity, PROMs, and subsequent therapies, aligned to HAS templates. Linkage to SNDS adds hospital/ambulatory costs, imaging, and procedures. This supports French-context utilities, resource use, and survival, which HAS prioritises over non-French datasets.

15) What ethical/privacy requirements govern RWE generation?

GDPR legal bases (public interest/legitimate interest), data minimisation, pseudonymisation/anonymisation.
DPIA (data protection impact assessment) and access controls.
National approvals: CNIL (France), NHS England DARS/IGARD (UK), provincial authorities (Canada), Datenschutz rules (Germany).
Transparency: lay summaries; opt-out mechanisms where applicable.

16) How can manufacturers design RWE acceptable to multiple HTA bodies?

Conduct early scientific advice with NICE/HAS/CADTH/IQWiG to align eligibility, endpoints, comparators, and confounding plans.
Use harmonised protocols with core outcomes (OS/PFS, irAEs, PROMs, utilities) and country-specific add-ons (resource use, tariffs).
Pre-specify analysis hierarchies (primary comparative design; sensitivity/QBA).
Ensure local data representation (e.g., French subset; UK SACT slice).
Plan for lifecycle submissions (initial + reassessment with maturing RWE).

17) How is RWE used in early access/managed entry agreements?

UK Cancer Drugs Fund (CDF): mandates RWE collection to resolve immature survival or uncertain subgroups before routine commissioning.
France (suivi post-inscription): RWE collection post-LPPR/listing to confirm SMR/ASMR and calibrate price.
Agreements often embed stopping rules, outcome-based triggers, or caps informed by RWE.

18) How does RWE quantify health-related quality of life (HRQoL) in melanoma?

Through PROMs administered in registries/clinics: EQ-5D-5L (mapped to utilities), EORTC QLQ-C30, QLQ-MC, and FACT-Mel. Methods:

Mixed models for repeated measures.
State-dependent utilities by line of therapy and irAE grade.
Mapping from disease-specific scales to EQ-5D where direct utilities are missing.
Utilities and irAE disutilities materially shift QALYs in cost-utility models.

19) What are the biggest methodological challenges in melanoma RWE?

Selection and channeling bias (fitter patients to ICIs/combos).
Immortal-time and time-varying confounding (responders stay longer on therapy).
Missingness (stage/ECOG/PROMs), misclassification (coding), left truncation.
Comparator drift as standards evolve (e.g., adjuvant ICI to metastatic sequencing).
Mitigation: target-trial emulation, G-methods, robust sensitivity/QBA, prospective registry designs.

20) What is the future of RWE in oncology HTA and reimbursement?

A hybrid evidence paradigm: RCT backbone + pragmatic trials + registry-based randomised/cluster designs and high-fidelity RWE integrated through:

Common data models and federated analytics (privacy-preserving linkage).
Real-time outcome monitoring for adaptive pricing and indications.
Harmonised EU HTA (Reg. 2021/2282) with country-level economic modules.
Standardised QBA and causal reporting checklists becoming expected in submissions.
Melanoma will remain the test-bed given durable responses, biomarker evolution, and high budget impact.