replying to Grannis Jr. et al. Nature Communications https://doi.org/10.1038/s41467-022-31419-9 (2022)
In the Mathios et al. study1, we demonstrated that a fragment-based cell-free DNA classifier (DELFI) evaluated on the prospectively collected participants in the LUCAS cohort could distinguish between patients with and without lung cancer and that the performance extended to an external validation dataset. We also demonstrated that DELFI could delineate patients with small cell lung cancer from other lung cancers and that the overall score was independently predictive of patient prognosis after adjustment for clinical factors including age, cancer stage, histological subtype, and treatment (Supplementary Fig. 11 in previous study).
In his comment on our study, Dr. Grannis raised the concern that our analytic cohort included many individuals with symptoms suggestive of lung cancer, rather than a symptom-free screening cohort. We highlighted this divergence in our paper too, but several points might alleviate this concern. First, we note it is common to initially evaluate cancer screening modalities in symptomatic individuals. The mammogram, colonoscopy, and prostate specific antigen test were all originally used for the evaluation of symptoms and signs of disease2,3,4. Additionally, according to the United States Preventative Task Force (USPTF) recommendation, individuals eligible for computed tomography (CT) screening include those with significant smoking history. The majority of our patients who were symptomatic had dyspnea and cough, symptoms common in the eligible population of heavy smokers that would be included in a screening study. We also included individuals with a prior history of cancer and comorbidities such as autoimmune and inflammatory conditions, traits that would be present in real-world screening populations.
We agree with Dr. Grannis regarding the proper sequence of steps for the development of an early detection test and re-emphasize that our analyses were proof of concept, building on another set of fragment based analyses we previously described across multiple cancer types5. Further development of a DELFI classifier using screening cohorts is still needed and Dr. Grannis points to a study that has already been launched: the L101 study aims “to train and test classifiers for lung cancer detection using the DELFI assay”, and focuses on individuals eligible for CT screening6.
Dr. Grannis expresses curiosity regarding the size of nodules in our study, based on his theory that the size of the nodule is itself a predictor of the clinical benefit of detecting it. Our study captured nodule size using the widely recognized TNM staging system and more detailed nodule dimensional measurements were not available for the cohorts we analyzed. Future analyses with cohorts with detailed radiographic annotation of lung nodules would be of value to assess the performance of the DELFI approach in detecting very small lung cancer lesions.
Dr. Grannis notes that a biomarker assay could harm patients by missing some sub-centimeter lung cancers. We agree that any biomarker must be assessed for its ability to identify patients with the earliest stage of disease, and that high false negative rates for small lesions would be a serious concern. However, the importance of the false negative rate of any biomarker for lung cancer should be balanced with the current low uptake of CT screening. A study by Jemal et al.7 cited a rate of 3.3–3.9% of eligible individuals undergoing CT imaging. A blood based assay could ameliorate this problem if two somewhat safe assumptions are met: a blood test would be adopted more easily by physicians as a first assessment for lung cancer than CT screening has been and, following a positive test, a CT scan would be ordered with high frequency.
The potential harms of CT screening include the frequency of false positives, the anxiety and cost associated with the test, and the high levels of exposure to ionizing radiation even at low doses. The benefits derive almost exclusively from the identification of patients with undiagnosed lung cancers who can then receive treatment. A blood-based biomarker could favorably shift the risk-benefit tradeoff of CT screening by identifying the subset of eligible individuals who are most likely to have lung cancer. Each positive CT scan would be more likely to identify a treatable cancer (higher positive predictive value), favorably shifting the true positive to false positive ratio and sparing many of those who are unlikely to have lung cancer from the harms associated with CT scanning.
This potential use of blood based early detection as a means of minimizing risks associated with CT screening is in line with the improvements in lung scan interpretation we have witnessed over the past decade. The Lung-RADS algorithm for CT scan evaluation and management, when applied to the NLST cohort retrospectively, reduces the sensitivity for lung cancer from 93.5% to 84.9%, but provides more than a 50% reduction in the baseline false positive rate (26.6–12.8%)8. I-ELCAP investigators, also in an analysis of NLST data, concluded that “higher thresholds of nodule size should be considered and prospectively evaluated”9. At the extremes the authors contemplate the impact of increasing the threshold for nodule evaluation from 5 mm to 9 mm. The higher threshold for nodule evaluation would more than halve the false positive rate (10.5–4.1%) but lead to delays in lung cancer diagnosis of greater than 9 months for 9.9% of patients by reducing the sensitivity of CT screening.
In conclusion, Mathios et al. demonstrate that non-invasive lung cancer detection is feasible through evaluation of genome-wide cfDNA fragmentation. Two larger trials, focused on CT-eligible screening populations are ongoing (NCT04825834, DELFI-L101 and NCT05306288, CASCADE-LUNG) with the goal of training and validating a classifier for early detection of lung cancer. Given the low depth of whole genome sequencing utilized by the DELFI approach, a test based on this method could provide an affordable and highly accessible avenue for population-scale lung cancer screening.
References
Mathios, D. et al. Detection and characterization of lung cancer using cell-free DNA fragmentomes. Nat. Commun. 12, 5060 (2021).
Bohlman, T. W. et al. Fiberoptic pansigmoidoscopy. An evaluation and comparison with rigid sigmoidoscopy. Gastroenterology 72, 644–649 (1977).
James, W. B. & Irvine, R. W. Mammography in management of breast lesions. Br. Med. J. 4, 655–657 (1969).
Stamey, T. A. et al. Prostate-specific antigen as a serum marker for adenocarcinoma of the prostate. N. Engl. J. Med. 317, 909–916 (1987).
Cristiano, S. et al. Genome-wide cell-free DNA fragmentation in patients with cancer. Nature 570, 385–389 (2019).
Force, U. S. P. S. T. et al. Screening for lung cancer: US preventive services task force recommendation statement. JAMA 325, 962–970 (2021).
Jemal, A. & Fedewa, S. A. Lung cancer screening with low-dose computed tomography in the United States-2010 to 2015. JAMA Oncol. 3, 1278–1281 (2017).
Pinsky, P. F. et al. Performance of lung-RADS in the national lung screening trial: a retrospective assessment. Ann. Intern. Med. 162, 485–491 (2015).
Yip, R. et al. The impact of the regimen of screening on lung cancer cure: a comparison of I-ELCAP and NLST. Eur. J. Cancer Prev. 24, 201–208 (2015).
Author information
Authors and Affiliations
Contributions
D.M., P.E.B., J.A.P., R.B.S., and V.E.V. all contributed to writing the paper.
Corresponding authors
Ethics declarations
Competing interests
D.M., J.A.P., R.B.S. and V.E.V. are inventors on patent applications submitted by Johns Hopkins University related to cell-free DNA for cancer detection. P.B.B. is an employee of Delfi Diagnostics, scientific advisor for EQRx and director of OncoHealth. J.A.P. and R.B.S. are founders of Delfi Diagnostics. V.E.V. is a founder of Delfi Diagnostics, serves as on the Board of Directors and as a consultant for this organization, and owns Delfi Diagnostics stock, which is subject to certain restrictions under university policy. Additionally, Johns Hopkins University owns equity in Delfi Diagnostics. V.E.V. divested his equity in Personal Genome Diagnostics (PGDx) to LabCorp in February 2022. V.E.V. is an inventor on patent applications submitted by Johns Hopkins University related to cancer genomic analyses and cell-free DNA for cancer detection that have been licensed to one or more entities, including Delfi Diagnostics, LabCorp, Qiagen, Sysmex, Agios, Genzyme, Esoterix, Ventana and ManaT Bio. Under the terms of these license agreements, the University and inventors are entitled to fees and royalty distributions. V.E.V. is an advisor to Danaher, Takeda Pharmaceuticals, and Viron Therapeutics. These arrangements have been reviewed and approved by the Johns Hopkins University in accordance with its conflict-of-interest policies.
Peer review
Peer review information
Nature Communications thanks the anonymous reviewer(s) for their contribution to the peer review of this work.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Mathios, D., Bach, P.B., Phallen, J.A. et al. Reply to: Limitations of molecular testing in combination with computerized tomographic for lung cancer screening. Nat Commun 13, 3892 (2022). https://doi.org/10.1038/s41467-022-31420-2
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41467-022-31420-2
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.
