@article{discovery10206230,
            note = {This version is the author-accepted manuscript. For information on re-use, please refer to the publisher's terms and conditions.},
          volume = {388},
           title = {Screening for multiple cancers: evaluation must go beyond aggregate measures},
           month = {February},
         journal = {The BMJ},
            year = {2025},
       publisher = {BMJ Publishing Group},
            issn = {0959-535X},
             url = {https://doi.org/10.1136/bmj-2024-081098},
        abstract = {Tom Callender and colleagues argue that aggregate outcomes amongst groups of cancers
are insufficient for the evaluation of multi-cancer detection tests for cancer screening and
that research providing cancer-specific outcomes will be necessary.
Multi-cancer detection tests (MCDs) using blood-based biomarkers for the concurrent
screening of multiple cancer types in asymptomatic individuals are currently being trialled in
the UK and elsewhere [1]. This has led to talk of their adoption in early cancer population
screening programmes.
Considerable attention has been paid to the potential benefits that MCDs could present by
improving cancer survival, particularly that of rarer cancers, through early detection.
Balanced against these benefits has been a focus on the possible harms from false
positives; inaccurate predictions of, or indeed inability to, predict the cancer site; and cases
where a cancer is detected by an MCD but is too small to be seen on imaging [2,3].
Most discussion has considered the overall - aggregate - benefits and harms of an MCD
programme. However, cancers behave remarkably differently both between individuals and
cancer types. Despite this, the benefits and harms accruing to each type of cancer with
strategies that uses a single multi-cancer screening test has had relatively little emphasis.
In this analysis, we discuss how differences in disease natural history could affect one
potential harm, overdiagnosis, when screening for multiple cancers simultaneously. We
argue that a focus on aggregate, all-cancer, metrics to evaluate MCDs will obscure
differences between cancers that are clinically and policy relevant. Building on this, we
discuss the benefits and challenges of approaches that balance the translation of a
promising MCD screening programme against the further research that may be required to
inform robust policy.},
          author = {Callender, Thomas and Payne, Katherine and Pashayan, Nora and Mackie, Anne}
}