Regarding “Survival Outcomes in Asymptomatic Patients with Normal Conventional Imaging but Raised Carcinoembryonic Antigen Levels in Colorectal Cancer Following Positron Emission Tomography‐Computed Tomography Imaging”

Computed tomography‐positron emission tomography (CT‐PET) scanning detects occult metastases in patients with raised carcinoembryonic antigen (CEA) during follow‐up for treated primary colorectal cancer (CRC). The study by Khan et al. confirms the well‐known sensitivity of this diagnostic technique but does not provide convincing objective data that such a screening policy is worthwhile in terms of improving survival of these patients [1]. Although the survival outcomes seemed to the authors to be longer than they would have expected, this is not a randomized study and so is subject not only to lead time bias (as they acknowledge) but also to selection bias, immortal time bias, and affirmation bias.

In their introduction, the authors refer to the systematic review and meta‐analysis by Pita‐Fernandez et al. [2], which appeared to show a survival benefit from increased diagnostic (including CEA measurement and CT scanning) surveillance of patients with CRC. We have recently published a similar systematic review, and our meta‐analysis, which included more recent data and excluded randomized trials with a high risk of bias and those with strategies only able to detect local recurrence, showed no significant survival benefit (hazard ratio 0.98, 95% confidence interval 0.87–1.11) [3]. There is therefore no secure evidence that imaging surveillance to detect occult metastatic disease improves overall survival.

Khan et al. also refer to “the common belief…that aggressive management of oligo‐metastatic disease in CRC leads to better survival outcomes.” It is indeed a widespread belief that has started to influence the management of other common cancers. But it is just a belief, albeit a strong one, based on the findings of observational studies, not on the results of any randomized trials, and contrary to the conventional understanding that in patients with carcinoma, metastases reflect a systemic rather than localized problem.

We quite understand the well‐meaning intentions of those who advocate increasing diagnostic surveillance of apparently cured cancer patients as is implied by this paper. However, in this era of mounting cost pressures on healthcare and, in the U.K. especially, of pressure on diagnostic imaging services, this policy should perhaps be considered more critically and the results of this study not interpreted as showing any real benefit from CT‐ PET imaging in this context.

Disclosures

The authors indicated no financial relationships.

References

Author notes