UK National Screening Committee position statement on evidence required for multicancer detection tests Authors: Sarah Batson, Sian Taylor-Phillips, Anne Mackie, Rosalind Given-Wilson, Steve Halligan, Allan Hackshaw, Tom Callender, Chris Hyde, Katherine Payne, Cristina Visintin, David Thompson, D Gareth Evans, Mike Richards, Bethany Shinkins Published: 2026 ABSTRACT Bethany Shinkins and colleagues outline a consensus set of principles and evidence requirements for evaluating multicancer detection tests in screening programmes to inform UK policy Multicancer detection (MCD) tests, which can identify multiple cancers from a single sample (eg, blood, urine, stool, or imaging), are a potential new approach to cancer screening.1 2 Several MCD tests are being developed, each at different stages of development and evaluation.3 4 5 Notably, in England, the NHS Galleri trial, a randomised controlled trial with 142 000 participants, is evaluating a blood based MCD test in a screening context.6 In the UK, the UK National Screening Committee (UK NSC) makes recommendations to the government regarding whether a new screening programme should be introduced. A topic can enter their horizon scanning process through multiple routes—for example, routine reviews of existing recommendations, the open call for screening proposals,7 or referrals from external bodies.8 The UK NSC then applies a structured evidence review process, progressing from evidence mapping to more detailed evidence summaries. When making recommendations, the UK NSC applies a multicriteria deliberative approach, considering factors such as disease burden, test performance, potential benefits and harms, cost effectiveness, and feasibility of delivery within the health system.9 This framework is designed to ensure that screening programmes deliver net population benefit and represent good value for public investment, while recognising the importance of minimising harms. Once implemented at scale, screening programmes can be difficult to modify or disinvest from,10 making it important to generate robust evidence before implementation. Organised screening programmes are currently in place for a few cancers, including breast, cervical, colorectal, and lung cancer.11 Although MCD tests are available through private purchase in some places, no country has yet implemented MCD tests within an organised screening programme.9 Generating robust evidence is inherently challenging for screening programmes, and evaluating MCD tests is further complicated by their aim to detect multiple cancers simultaneously. The most reliable way to determine whether cancer screening reduces mortality is through adequately powered and well designed randomised trials comparing outcomes in those offered screening and those not offered screening. To support consistent, transparent, and proportionate decision making in this rapidly evolving area, this position statement, intended for manufacturers, researchers, and funders, sets out the principles and evidence required to inform a decision about an MCD screening programme. These principles span key aspects of screening evaluation discussed in this statement, structured using the population, intervention, comparator, and outcome (PICO) framework.12 We have intentionally not been prescriptive about study design as the different types of evidence we outline could be generated using a vast array of different approaches and we do not want to restrict methodological innovation in this space. How this position statement was developed An independent taskforce was established by invitation from the UK NSC secretariat to develop this position statement. The taskforce included clinicians, statisticians, health economists, behavioural scientists, epidemiologists, an ethicist, public health consultants, and patient and public voice representatives. The taskforce met regularly between July 2024 and October 2025 to review evidence, discuss key issues, and reach consensus on the final statement. First, the methodological literature was reviewed to identify aspects of evidence generation and evaluation that may be distinct or particularly challenging for MCDs. We used the PICO framework12 to group emerging themes from this review to structure subsequent discussion. For each issue identified, the taskforce was presented with viewpoints from the literature and discussed the extent to which these were agreed with or contested using a deliberative process. Finally, position statements were drafted based on these discussions, presented to the taskforce, and refined iteratively during meetings and through written feedback between meetings, until the final wording was agreed. Defining the use case for MCD screening When evaluating a potential MCD screening programme, the UK NSC requires a clear description of the specific way in which the test would be used in practice. Examples of possible use cases include a population-wide programme with eligibility based on age and, possibly, sex; a programme targeted to one or more defined high risk subgroup(s); and as an add-on within an existing single cancer programme, where the MCD test is offered as part of the existing screening pathway. The use case definition should describe how, where, and for whom the MCD test is intended to be used in the screening programme, and the proposed pathway from screening to diagnosis and follow-up (table 1). Table 1Minimum elements of multicancer detection (MCD) screening programme use caseComponentDescriptionCancers being screened forThe specific cancers the programme aims to detectThe population to be screenedThe target population, including whether the programme is population based (eg, all individuals within a specified age group and sex) or targeted (eg, higher risk)Identification and invitation methodsHow eligible individuals will be identified and invited for screening (eg, via primary care health records, population registers, or integration with existing screening programmes), noting that this may differ across screening contexts within each devolved nation in the UKThe MCD test to be usedThe proposed test, including key characteristics such as detection method (eg, cell-free DNA), sample type, and whether the test predicts cancer site of originData analysis, storage and integrationHow test data will be stored and integrated into patient records and clinical systemsFollow-up investigationsThe specific investigations or clinical pathway triggered by a positive test resultScreening frequency and timingThe proposed screening interval Minimum elements of multicancer detection (MCD) screening programme use case Population Interaction with existing screening programmes Introducing an MCD screening programme could potentially influence participation in existing screening—for example, by shaping perceptions of relevance, displacing attention, or generating false reassurance after a negative MCD test result. Understanding the interaction between MCD and existing screening pathways is therefore important for informing policy decisions. When MCD screening is being considered as a standalone programme, evidence is needed to understand its potential effect on uptake, adherence, and engagement with existing cancer screening. Targeted screening in high risk populations An MCD screening programme may be proposed for targeted or high risk populations, such as people with genetic predispositions or raised risk profiles. Targeted screening may also be considered if randomised trials of population level MCD screening do not show clinical or cost effectiveness but prespecified subgroup analyses indicate benefit within a clearly defined high risk group. In such scenarios, the evidence should be sufficiently powered to support robust conclusions for the targeted population, rather than relying solely on exploratory or post hoc analyses. Any proposal should be supported by a clear rationale and appropriately stratified evidence showing that the MCD test performs adequately in the target population, that the expected benefits outweigh the harms, and that the programme is likely to be cost effective. Proposals should also show that the target population can be identified reliably and reached in practice, which may require additional evidence on risk stratification tools, data infrastructure, and implementation pathways. Risk prediction models used to define eligibility for risk stratified screening should be developed independently of the trials evaluating the effectiveness of that screening approach. When targeted screening is being considered, early engagement with the UK NSC research and methodology group is encouraged to clarify whether responsibilities may be shared with, or lie with, other bodies. Embedding MCD tests into existing programmes MCD tests could also be incorporated into existing cancer screening programmes, which would constitute a major modification and should be assessed through the UK NSC’s established process for evaluating such changes. This process requires evidence that the modified cancer screening programme continues to meet the UK NSC’s screening criteria, considering both the new and existing components.9 Key evidence requirements include how the addition of the MCD test affects the overall clinical effectiveness and cost effectiveness of the programme; whether the addition of the MCD test alters uptake of the existing screening modality, compliance with follow-up, or participant understanding; the safety and acceptability of offering multiple screening modalities in a single encounter; and operational feasibility and impact on programme delivery. Proposals to embed MCD tests within an existing programme should be supported by evidence generated under realistic service conditions. This may require implementation studies, in addition to any trial based evidence, to assess how the combined tests perform in practice. Intervention Screening strategy (timing and frequency) Determining the optimal MCD screening strategy, including frequency and timing of invitations, is a fundamental component of intervention design. Unlike screening for individual cancers, an MCD programme spans multiple malignancies with differing epidemiology, tumour biology, and progression rates. As such, there is unlikely to be a single testing interval optimal for all targeted cancers. Practical decisions about frequency will therefore need to balance clinical effectiveness, cost effectiveness, health system capacity, and interactions with existing screening programmes. End-to-end randomised trials of MCD screening with repeated testing and long term follow-up are likely to be rare and resource intensive. Trials may therefore incorporate more frequent testing than expected in practice, to evaluate different intervals and to support economic modelling of different strategies. However, this complicates evidence interpretation; estimates of clinical effectiveness (eg, mortality benefit) may apply only to the specific schedule evaluated, and extrapolation to other frequencies must be supported by robust modelling and interpreted with caution. Economic modelling, grounded in natural progression of the disease and empirical data on clinical effectiveness, resource use, and system constraints will be key when evaluating different MCD screening approaches. Assumptions underpinning such models should be transparent, justified, and clearly linked to evidence, including the intervals at which effectiveness data were collected. Evidence driven selection of cancers for screening Not all detectable cancers will necessarily be suitable for inclusion in a screening programme; earlier detection may not improve outcomes and may cause harms such as overdiagnosis. It is therefore important to distinguish between the technical capabilities of the test and the cancers for which screening is expected to deliver net benefit. An MCD screening programme may include fewer cancers than were evaluated in a trial if evidence suggests the harms of screening for certain cancers outweigh the benefits. However, the feasibility of excluding specific cancers may depend on the design and reporting structure of the test. To support such decisions, results should be reported separately for each cancer, as well as in aggregate. This transparency is essential for understanding which cancers contribute meaningfully to the overall balance of benefits and harms, and for designing a programme that maximises population benefit. Diagnostic work-up after a positive result An MCD screening programme is defined not only by the test itself but also by the diagnostic pathway that follows a positive result. The potential clinical and system effect of an MCD programme will depend heavily on the investigations triggered by a positive result. Studies evaluating clinical effectiveness should collect evidence that reflects real world care pathways as closely as possible. Detailed data on diagnostic work-up should be collected and reported, including (but not limited to) the following: The range and sequence of subsequent investigations (eg, imaging, blood tests, biopsies) The range and sequence of subsequent investigations (eg, imaging, blood tests, biopsies) Any variation in diagnostic pathways based on the predicted cancer site of origin, the possibility of cancer recurrence, clinical judgment, or local protocols Any variation in diagnostic pathways based on the predicted cancer site of origin, the possibility of cancer recurrence, clinical judgment, or local protocols Time to definitive diagnosis, including cases where no cancer is ultimately found Time to definitive diagnosis, including cases where no cancer is ultimately found The proportion of people having invasive or high burden procedures, including those resulting from false positive results; their associated healthcare resource use; and any associated psychological impact from how unresolved or indeterminate findings are managed. The proportion of people having invasive or high burden procedures, including those resulting from false positive results; their associated healthcare resource use; and any associated psychological impact from how unresolved or indeterminate findings are managed. The UK NSC will expect clearly defined diagnostic work-up, justification of feasibility within NHS capacity, and consideration of any associated psychological and physical harms. Handling unresolved cases and false positive results A distinctive feature of an MCD test is that a positive result may indicate cancer in multiple possible sites, creating diagnostic uncertainty. In some cases, no cancer may be found despite subsequent investigations. This uncertainty is likely to be distressing, particularly if people are left without a clear explanation of whether a cancer was missed or the result was a false positive. These unresolved cases therefore raise important questions about follow-up, communication, and longer term management. The UK NSC requires evidence on How often positive MCD test results do not lead to a cancer diagnosis (in the short and longer term), How often positive MCD test results do not lead to a cancer diagnosis (in the short and longer term), Protocols proposed to manage people with unresolved findings, Protocols proposed to manage people with unresolved findings, Psychological impact and acceptability of uncertainty for screening participants, and Psychological impact and acceptability of uncertainty for screening participants, and Clinical frameworks, including ethical considerations, for discharging or re-evaluating patients over time. Clinical frameworks, including ethical considerations, for discharging or re-evaluating patients over time. Protocols should be clearly defined, proportionate, and feasible to implement at scale. Evidence should show that harms such as anxiety, overinvestigation, or delayed diagnosis, are minimised. Approaches to unresolved cases should also be communicated before testing to support informed choice about participation. Test iteration and version control MCD tests are often based on complex, evolving technologies that may be updated during or after the evidence generation process. Updates can affect test performance, including sensitivity and specificity for cancer detection and the accuracy of predicting the site of origin. As such, the specific test version evaluated should be clearly defined and documented. The UK NSC expects evidence to correspond to a stable and well defined MCD test version. Any modifications should be transparently reported, and the effect of changes on test performance and pathway outcomes clearly assessed. Evidence generated using an earlier test version may not be sufficient to inform policy decisions if substantial modifications have been made. Sponsors and developers should provide a version control history and evidence that updated versions are analytically and clinically equivalent—if not superior. When substantial changes occur, additional validation or bridging studies may be required. Communication of results and informed choice Effective communication is a fundamental component of any screening intervention. For MCD screening, the complexity of test results, including uncertainty about cancer presence, type, or location, means that clear, accessible information is especially important to support informed choice and appropriate follow-up. People offered MCD screening should be given information about What the test can and cannot detect, What the test can and cannot detect, What a negative result means (including making it explicit that a negative result does not rule out cancer and should not influence participation in existing recommended screening programmes), What a negative result means (including making it explicit that a negative result does not rule out cancer and should not influence participation in existing recommended screening programmes), What a positive result means (including being clear that it is possible that no cancer will be found after a positive MCD result), What a positive result means (including being clear that it is possible that no cancer will be found after a positive MCD result), The uncertainty around predicting the cancer site of origin, and The uncertainty around predicting the cancer site of origin, and The potential for multiple further investigations. The potential for multiple further investigations. Informed choice is fundamental and should be supported by evidence based materials that are clear, accessible, and acceptable to diverse populations. Communication materials and processes should have been shown to enable informed decision making, support psychological wellbeing, and be acceptable across different demographic groups. This should include pre-test information, communication of results, and follow-up discussions and should be evaluated empirically where possible (eg, user testing, surveys, or qualitative research embedded within trials or pilots). Special attention should be paid to ensuring that people with lower health literacy or limited English proficiency are not disadvantaged in the screening process. Comparators Identifying the most appropriate comparator for randomised trials The comparator in a trial should reflect current or expected future practice in the UK for each cancer included in the screening programme. For cancers with an existing screening programme, the comparator will be that current programme, across all relevant cancers, including all applicable screening pathways. For cancers without an established screening programme, the appropriate comparator is no screening. Changes in UK clinical practice after an MCD trial was conducted may affect the relevance of the comparator and interpretation of trial findings. Comparators should align with how the MCD test is intended to be used in practice and enable meaningful interpretation of its incremental value. This supports assessment of its effect on clinical outcomes, cost effectiveness, and integration with existing care pathways. Although the UK NSC will prioritise evidence from current UK practice, studies from other countries or health systems may be considered if results are generalisable to the UK context. Because MCD screening spans multiple cancers, trial design should capture participants’ engagement with existing screening programmes and relevant risk factors. This information is critical to interpreting results and estimating the additional benefit of an MCD screening programme in a real world setting. Use of economic modelling to supplement limited comparator arms Randomised trials of MCD screening programmes are large, complex, and expensive, and it is not feasible to reflect all possible relevant comparators. This means that some important comparisons may not be assessed directly in a trial. To address this, economic modelling using decision-analysis frameworks should be used, where possible, to explore a wider range of comparators and estimate long term clinical and cost-effectiveness outcomes. By combining trial results with other relevant data sources, economic modelling can provide a structured and transparent approach to comparing MCD screening with current practice and potential alternative strategies, including modelling of potential harms such as false positive results, associated burden of care, and effects on healthcare services, including primary care. Economic modelling is therefore a crucial tool for informing screening policy and understanding the full value of MCD screening programmes. Outcomes Multicriteria decision making The UK NSC uses a deliberative process to decide whether to recommend a screening programme, based on a broad range of criteria.9 Although reduction in disease specific mortality is an important outcome, it is not the sole determinant. Decisions are based on the overall balance of benefits and harms, accounting for a broad set of outcomes and contextual factors. Trials are an essential starting point for assessing whether a screening programme should be considered. They can generate high quality evidence regarding performance and potential effects but are unlikely to provide all the evidence needed for a policy recommendation. The value of a trial lies not only in whether it shows a mortality benefit, but in how it contributes to the wider assessment of whether the benefits of screening outweigh the harms. Relevant outcomes The UK NSC will consider multiple outcomes when evaluating MCD screening including, but not limited to, those listed in box 1. Trials and other studies should be designed to measure these outcomes directly or collect data to inform their estimation through complementary analyses. Registry linkage or other follow-up mechanisms may also be used to capture cancers occurring after both positive and negative test results. Examples of outcomes relevant to evaluating an MCD screening programme Safety Safety of the screening test Safety of the screening test Safety of any follow-up diagnostic tests Safety of any follow-up diagnostic tests Safety of any treatment given as a result of screening Safety of any treatment given as a result of screening Participation, equity, and acceptability Participation in MCD screening and adherence to whole pathway Participation in MCD screening and adherence to whole pathway Inequity in uptake, adherence, and follow-up Inequity in uptake, adherence, and follow-up Inequity in access Inequity in access Acceptability of screening, diagnostic, and treatment pathways Acceptability of screening, diagnostic, and treatment pathways Test performance Diagnostic accuracy Diagnostic accuracy Accuracy of prediction cancer site of origin Accuracy of prediction cancer site of origin Test failure rate Test failure rate Rate of unresolved positive cases (screen test positive but cancer never found) Rate of unresolved positive cases (screen test positive but cancer never found) Clinical effectiveness Cancer specific and all-cause mortality Cancer specific and all-cause mortality Health related quality of life Health related quality of life Morbidity Morbidity Late stage incidence of cancer (interim) Late stage incidence of cancer (interim) Rates of interval cancers Rates of interval cancers Harms and unintended consequences Decreased uptake of existing screening programmes Decreased uptake of existing screening programmes Fewer people seeking help about their symptoms Fewer people seeking help about their symptoms Late stage presentation of cancers not screened for Late stage presentation of cancers not screened for Overdiagnosis or overtreatment Overdiagnosis or overtreatment Psychological harms Psychological harms System and budget impact Cost and opportunity cost Cost and opportunity cost Effect on use of constrained resources (eg, primary care, imaging. and treatment waiting times) Effect on use of constrained resources (eg, primary care, imaging. and treatment waiting times) Management of incidental findings (from MCD test and follow-up diagnostic tests) Management of incidental findings (from MCD test and follow-up diagnostic tests) Disaggregated outcomes by cancer Because MCD screening targets multiple cancers, aggregate trial outcomes may mask important variation in benefit and harm across cancer types. The UK NSC recognises that trials may not be powered for rarer cancers; however, in addition to overall outcomes, cancer specific outcomes should be reported where possible, even if underpowered. These data support interpretation of the overall results, help identify which cancers contribute most to net benefit, and inform economic modelling of cancer specific benefits and harms. The UK NSC may also request aggregate outcomes excluding certain cancers (for example, cancers already covered by existing screening programmes such as bowel or lung) to better understand the incremental value of MCD screening. Surrogate outcomes Given the size, long follow-up, and high cost of randomised trials of screening, interim endpoints may be useful for the UK NSC to decide whether further evaluation is necessary. A separate position statement on the general use of surrogate outcomes for cancer screening trials is also published in The BMJ (doi:10.1136/bmj-2026-629407).13 On the basis of current evidence, reductions in the incidence of late-stage (stage III and IV) cancer are considered the most promising surrogate outcome for cancer specific mortality, which is a key outcome of interest for the UK NSC. However, the magnitude of effect and poor prediction for some cancers remain critical considerations. A key concern when relying on surrogate endpoints, particularly relevant for MCD tests, is the potential for tests to preferentially detect cancers with more aggressive biology or micro-metastatic disease. This could weaken the relation between reductions in late stage (stage III and IV) incidence and cancer specific mortality, leading to an apparent reduction in late stage incidence without a corresponding improvement in cancer specific mortality. Given the wide range of outcomes of interest to the UK NSC, the likelihood of substantially underpowered outcomes for rarer cancers, and the substantial implementation issues needing consideration before full programme rollout, further (potentially in-service) evaluation is likely to be needed if promising interim trial outcomes are reported. Conclusion Using an MCD test within a screening programme may improve cancer outcomes by enabling earlier diagnosis across multiple tumour types. The UK NSC recognises this potential and is awaiting robust evidence on whether an MCD screening programme delivers meaningful net benefit at reasonable cost. If questions remain unanswered by this position statement, we strongly encourage engagement with the UK NSC research and methodology group.14 This position statement emphasises the complexity of MCD screening and the need for an evidence base that reflects that complexity (box 2). A guiding principle for the UK NSC is to welcome innovation while ensuring that any recommendation to introduce a screening programme is based on rigorous evaluation. Any programme should demonstrate net benefit, cost effectiveness, and feasibility within the UK health system. Main considerations in evaluating MCD screening The UK NSC applies a multicriteria deliberative approach, considering factors such as disease burden, test performance, potential benefits and harms, cost effectiveness, and feasibility of delivery within the health system: · A clearly defined use case for MCD screening is essential · A clearly defined use case for MCD screening is essential · A portfolio of evidence is likely to be required with evidence that fully addresses the population, intervention, comparator, and outcomes · A portfolio of evidence is likely to be required with evidence that fully addresses the population, intervention, comparator, and outcomes · Refer to the UK NSC position statement on the use of surrogate outcomes in cancer screening trials (doi:10.1136/bmj-2026-629407)13 · Refer to the UK NSC position statement on the use of surrogate outcomes in cancer screening trials (doi:10.1136/bmj-2026-629407)13 · Economic modelling will be important for synthesising evidence and exploring uncertainties; transparency in model development and reporting will be important · Economic modelling will be important for synthesising evidence and exploring uncertainties; transparency in model development and reporting will be important · Engage with the UK NSC research and methods group early (www.gov.uk/government/publications/uk-nsc-rmg-submitting-research-requests) · Engage with the UK NSC research and methods group early (www.gov.uk/government/publications/uk-nsc-rmg-submitting-research-requests) Key messages Evaluation of multicancer detection (MCD) tests for screening is inherently more complex than for single cancer tests Evaluation of multicancer detection (MCD) tests for screening is inherently more complex than for single cancer tests Evaluation frameworks must therefore reflect the diversity of cancers involved, including differences in natural progression of disease, existing screening pathways, and the balance of benefits and harms Evaluation frameworks must therefore reflect the diversity of cancers involved, including differences in natural progression of disease, existing screening pathways, and the balance of benefits and harms Clear specification of the intended use case is essential: evidence requirements will appropriately differ for population-wide screening, targeted high-risk screening, and use alongside existing single-cancer programmes Clear specification of the intended use case is essential: evidence requirements will appropriately differ for population-wide screening, targeted high-risk screening, and use alongside existing single-cancer programmes Policy decisions on MCD screening will be supported through a coherent portfolio of evidence, consistent with standard UK National Screening Committee practice; this should combine evidence from trials, qualitative research, implementation studies, and transparent economic modelling Policy decisions on MCD screening will be supported through a coherent portfolio of evidence, consistent with standard UK National Screening Committee practice; this should combine evidence from trials, qualitative research, implementation studies, and transparent economic modelling Decisions about MCD screening will be guided by a comprehensive assessment of benefits and harms, considering outcomes beyond early detection, including false positives, diagnostic burden, equity, acceptability, and cost-effectiveness Decisions about MCD screening will be guided by a comprehensive assessment of benefits and harms, considering outcomes beyond early detection, including false positives, diagnostic burden, equity, acceptability, and cost-effectiveness