Recommendations for Implementing Lung Cancer Screening with Low-Dose Computed Tomography in Europe
, Joseph Shemesh 9, John K. Field 10, Javier J. Zulueta 11,12, Denis Horgan 13, Lucia Fiestas Navarrete 5, Maurizio Valentino Infante 14, Pierluigi Novellis 2, Rachael L. Murray 15, Nir Peled 16, Cristiano Rampinelli 17, Gaetano Rocco 18, Witold Rzyman 19, Giorgio Vittorio Scagliotti 20, Luca Bertolaccini 22, Giuseppe Ferrante 26, Carlijn van der Aalst 8, Dario Consonni 29, Zaigham Saghir 33 and Giuseppe Pelosi 34,35,*add
Show full author list
Round 1
Reviewer 1 Report
This consensus paper investigated the different experience on screening programs using low dose CT and the recommendation to implement the screening starting from the published experience.
The conclusion are that a risk prediction model needs to be implemented.
LDCT protocol and quality control are essential
Please insert in the discussion definition of low dose and quality in lung CT screening.
Machine learning may help in detection but is still under validation please discuss.
Author Response
Reviewer #1:
Comments and Suggestions for Authors
This consensus paper investigated the different experience on screening programs using low dose CT and the recommendation to implement the screening starting from the published experience.
The conclusion is that a risk prediction model needs to be implemented.
LDCT protocol and quality control are essential
Please insert in the discussion definition of low dose and quality in lung CT screening.
Answer 1: we thank the reviewer for this suggestion. We added the definition of Low dose CT at page 15 of the revised text.
“One of the concerns of CT screening is related to the radiation exposure. There is no consensus on what level of radiation is considered ‘low-dose’. However, a low-dose lung cancer screening CT scan should be performed based on technical specifications ensuring that the quality of the screening and the radiation dose is in compliance with ACR-STR recommendations (https://www.acr.org/-/media/ACR/Files/Practice-Parameters/CT-LungCaScr.pdf?la=en).
Machine learning may help in detection but is still under validation please discuss.
Answer 2: according with the kind suggestion of the reviewer we added this sentence in the text at page 14
“For the future, machine learning may help in detection and characterization of screening detected nodules. Several publications have demonstrated the use of ML to characterize nodules from an image. Ardila et al (10) achieved 94.4% AUC performance on large number of cases of the National Lung Cancer Screening Trial, and validation sets. This creates an opportunity to optimize the screening process through IT assistance and case automation by enhancing advanced learning models in order to increase consistency and adoption of lung cancer screening worldwide” (Diego Ardila, Atilla P. Kiraly, Sujeeth Bharadwaj, et al. End-to-end lung cancer screening with three-dimensional deep learning on low-dose chest computed tomography. Nature Medicine 2019; 954–961.)
Reviewer 2 Report
...
The authors provided a recommendation for implementing lung cancer screening with low-dose computed tomography in Europe.
The manuscript is well structured and address the aspect in an objective and comprehensive approach. In generally, I recommend the manuscript for publication.
Meanwhile, there is an essential (minor) issue.
I recommend improving the discussion of radiobiological risks from LD-CT lung screening vs. the potential benefits from an early diagnosis of lung cancer. The authors provided only a short reflection of this essential aspect based on a publication by Rampinelli et al. [1] (reference 97, section 3.1. “Considerations on Volumetry and Doses“). Meanwhile, the response to the publication by Rampinelli et al. [1], including a critical discussion, illustrated the heterogenous and controversial positions regarding the specific aspects (in total five ‘rapid responses’). On the other hand, the safety driven discussion (e.g. with national regulatory authorities) reflect a standard in establishing screening (imaging) methodology using ionizing radiation.
The objective benefits from early diagnosis has to be discussed in comparison to the cohort attributed CT exposure. A consistent argumentation is essential and will provide support in establishing the recommended screening process.
Sincerely,
...
[1] Rampinelli C. et al., Exposure to low dose computed tomography for lung cancer screening and risk of cancer: secondary analysis of trial data and risk-benefit analysis. BMJ (Clinical research ed.) 2017, https://dx.doi.org/10.1136/bmj.j347.
Author Response
Reviewer #2:
The authors provided a recommendation for implementing lung cancer screening with low-dose computed tomography in Europe.
The manuscript is well structured and address the aspect in an objective and comprehensive approach. In generally, I recommend the manuscript for publication.
Meanwhile, there is an essential (minor) issue.
I recommend improving the discussion of radiobiological risks from LD-CT lung screening vs. the potential benefits from an early diagnosis of lung cancer. The authors provided only a short reflection of this essential aspect based on a publication by Rampinelli et al. [1] (reference 97, section 3.1. “Considerations on Volumetry and Doses”). Meanwhile, the response to the publication by Rampinelli et al. [1], including a critical discussion, illustrated the heterogenous and controversial positions regarding the specific aspects (in total five ‘rapid responses’). On the other hand, the safety driven discussion (e.g. with national regulatory authorities) reflect a standard in establishing screening (imaging) methodology using ionizing radiation.
The objective benefits from early diagnosis has to be discussed in comparison to the cohort attributed CT exposure. A consistent argumentation is essential and will provide support in establishing the recommended screening process.
[1] Rampinelli C. et al., Exposure to low dose computed tomography for lung cancer screening and risk of cancer: secondary analysis of trial data and risk-benefit analysis. BMJ (Clinical research ed.) 2017, https://dx.doi.org/10.1136/bmj.j347.
Answer 1: we are indebted with the reviewer for his interest in this important topic.
According to his/her suggestions we added a more detailed discussion on the topic at page 15 of the revised manuscript.
“A recent study evaluating the cumulative radiation exposure and cancer risk from low-dose CT screening, estimated a 0.05% additional risk of cancer after 10 years of screening and associated follow-up imaging. 99 However, different CT protocols and nodule management strategies in screening could lead to different levels of radiation exposure. The assessment of cancer risk from radiation is based on the linear no-threshold model 100 and on data collected from occupational studies and from atomic bomb survivors. The risks are thus based on models generated from studies on people exposed to high levels of radiation; therefore, the linear no-threshold model stands as a precautionary recommendation that follows a conservative approach.”
