To the Editor:
We read with interest the articles by Ueberroth et al
1
and Olivier et al- Ueberroth BE
- Marks LA
- Borad MJ
- Agrwal N
Multicancer early detection panels (MCEDs) in the primary care setting.
Am J Med. 2022 Mar 31; ([online ahead of print]) (S0002-9343(22)00242-X)https://doi.org/10.1016/j.amjmed.2022.03.006
2
about emergence of Multicancer Early Detection (MCED) in primary care. We applaud the effort to outline the salient issues about use of MCED tests with the goal of arming clinicians with factual knowledge about the rationale and performance of the tests, the key questions that patients may ask, and potential downsides of MCED. Both articles raise issues that require further clarification.While existing single cancer tests generally have good sensitivity for detecting stage I disease for individual cancer types, they have a lower all-cancer sensitivity than an MCED when considering all cancers as the denominator. The Circulating Cell-free Genome Atlas (CCGA) study showed that for 12 common cancers that account for two-thirds of US cancer deaths (anus, bladder, colon/rectum, esophagus, head and neck, liver/bile duct, lung, lymphoma, ovary, pancreas, plasma cell neoplasm, stomach), GRAIL's Galleri test had a sensitivity of 53% for stage I and II disease.
3
For ovarian and pancreatic cancers that effectively have a 0% early detection rate currently due to the lack of existing screening tests, Galleri detected 60% of stage I and II tumors.3
Ueberroth et al1
cite a post hoc analysis- Ueberroth BE
- Marks LA
- Borad MJ
- Agrwal N
Multicancer early detection panels (MCEDs) in the primary care setting.
Am J Med. 2022 Mar 31; ([online ahead of print]) (S0002-9343(22)00242-X)https://doi.org/10.1016/j.amjmed.2022.03.006
4
that suggests that the estimated positive predictive value (PPV) should be based on the prevalence of individual cancers, an assertion that is fundamentally incorrect for an MCED, where test performance is based on the aggregated prevalence of all cancers.5
Using a prespecified analysis, 2 studies,3
,6
including the interim results of one in an intended-use adult screening population (PATHFINDER),6
demonstrated that Galleri's PPV exceeds 40%, far higher than current single-cancer tests such as mammography7
and low-dose computed tomography for lung cancer,8
which have PPVs <10%.Olivier et al
2
question whether early detection is better and raise appropriate concerns about over-detection of nonlethal cancers. Chen et al9
demonstrated in the CCGA study3
,10
that, while cancers detected by Galleri had expected survival outcomes, those not detected had better than expected outcomes, clearly suggesting that over-detection of nonlethal cancers is less likely by this MCED. Additional research is needed to better define this area, but it is possible that by relying on circulating tumor DNA as the principal analyte, MCEDs in development will be more sensitive for cancers with lethal potential than for clinically insignificant cancers. While we share concerns about the potential harms that may result from false positive tests, there were several inaccuracies in the commentary related to the analysis of this issue. First, Olivier et al incorrectly imply that 200 million adults would be eligible to be screened by broad use of MCED, resulting in 1 million false positive results. Cancer screening in the United States is generally limited to those aged 50-79 years, defining an eligible population estimated at 107 million.11
Further, there is a 10-fold error in the calculation of the number of potential false positives based on a hypothetical sample size of 95,174 in a randomized trial: a 0.5% (0.005) rate of false positives would result in 476 false positives (95,174 × 0.005), not the 4758 estimated in the article. Relative to a collective estimated false positive rate of 7.5% for currently recommended screening tests11
(excluding prostate cancer, which is higher), the addition of an MCED such as Galleri, with a false positive rate of 0.005 (0.5%) and an estimated 26% reduction in cancer mortality,12
is likely to yield a substantial gain in the efficacy of screening. Finally, in the PATHFINDER study,6
testing was met with high levels of participant satisfaction regardless of result; most patients with a false positive test had only radiographic evaluation and not an invasive procedure, and there were no clinical adverse events associated with diagnostic evaluation of false positives.References
- Multicancer early detection panels (MCEDs) in the primary care setting.Am J Med. 2022 Mar 31; ([online ahead of print]) (S0002-9343(22)00242-X)https://doi.org/10.1016/j.amjmed.2022.03.006
- Multi-cancer screening tests: communicating about risks should be prioritized.Am J Med. 2022; 135: 413-415
- Clinical validation of a targeted methylation-based multi-cancer early detection test using an independent validation set.Ann Oncol. 2021; 32: 1167-1177
- Can a broad molecular screen based on circulating tumor DNA aid in early cancer detection?.J Appl Lab Med. 2020; 5: 1372-1377
- Universal cancer screening: revolutionary, rational, and realizable.NPJ Precis Oncol. 2018; 2: 23
- A prespecified interim analysis of the PATHFINDER study: performance of a multi‑cancer early detection test in support of clinical implementation.J Clin Oncol. 2021; 39: 3070
- Performance benchmarks for screening breast MR imaging in community practice.Radiology. 2017; 285: 44-52
- Applying the National Lung Screening Trial eligibility criteria to the US population: what percent of the population and of incident lung cancers would be covered?.J Med Screen. 2012; 19: 154-156
- Prognostic significance of blood-based multi-cancer detection in plasma cell-free DNA.Clin Cancer Res. 2021; 27: 4221-4229
- Sensitive and specific multi-cancer detection and localization using methylation signatures in cell-free DNA.Ann Oncol. 2020; 31: 745-759
- Estimating the population health impact of a multi-cancer early detection genomic blood test to complement existing screening in the US and UK.Br J Cancer. 2021; 125: 1432-1442
- Modeled reductions in late-stage cancer with a multi-cancer early detection test.Cancer Epidemiol Biomarkers Prev. 2021; 30: 460-468
Article info
Footnotes
Funding: None.
Conflicts of Interest: EAK is a consultant for GRAIL, Inc. MS is a consultant for GRAIL, Inc.
Authorship: All authors had access to the data and a role in writing the manuscript.
Identification
Copyright
© 2022 Elsevier Inc. All rights reserved.
