Assessment of lung cancer mortality reduction after chest X-ray screening in smokers: a population-based cohort study in Varese, Italy

Image quality of ultralow-dose chest CT using deep learning techniques: potential superiority of vendor-agnostic post-processing over vendor-specific techniques

OBJECTIVE

To compare the image quality between the vendor-agnostic and vendor-specific algorithms on ultralow-dose chest CT.

METHODS

Vendor-agnostic deep learning post-processing model (DLM), vendor-specific deep learning image reconstruction (DLIR, high level), and adaptive statistical iterative reconstruction (ASiR, 70%) algorithms were employed. One hundred consecutive ultralow-dose noncontrast CT scans (CTDI_vol; mean, 0.33 ± 0.056 mGy) were reconstructed with five algorithms: DLM-stnd (standard kernel), DLM-shrp (sharp kernel), DLIR, ASiR-stnd, and ASiR-shrp. Three thoracic radiologists blinded to the reconstruction algorithms reviewed five sets of 100 images and assessed subjective noise, spatial resolution, distortion artifact, and overall image quality. They selected the most preferred algorithm among five image sets for each case. Image noise and signal-to-noise ratio were measured. Edge-rise-distance was measured at a pulmonary vessel, i.e., the distance between two points where attenuation was 10% and 90% of maximal intravascular intensity. The skewness of attenuation was calculated in homogeneous areas.

RESULTS

DLM-stnd, followed by DLIR, showed the best subjective noise on both lung and mediastinal windows, while DLIR yielded the least measured noise (ps < .0001). Compared to DLM-stnd, DLIR showed inferior subjective spatial resolution on lung window and higher edge-rise-distance (ps < .0001). Additionally, DLIR showed the most frequent distortion artifacts and deviated skewness (ps < .0001). DLM-stnd scored the best overall image quality, followed by DLM-shrp and DLIR (mean score 3.89 ± 0.19, 3.68 ± 0.24, and 3.53 ± 0.33; ps < .001). Two among three readers preferred DLM-stnd on both windows.

CONCLUSION

Although DLIR provided the best quantitative noise profile, DLM-stnd showed the best overall image quality with fewer artifacts and was preferred by two among three readers.

KEY POINTS

• A vendor-agnostic deep learning post-processing algorithm applied to ultralow-dose chest CT exhibited the best image quality compared to vendor-specific deep learning algorithm and ASiR techniques. • Two out of three readers preferred a vendor-agnostic deep learning post-processing algorithm in comparison to vendor-specific deep learning algorithm and ASiR techniques. • A vendor-specific deep learning reconstruction algorithm yielded the least image noise, but showed significantly more frequent specific distortion artifacts and increased skewness of attenuation compared to a vendor-agnostic algorithm.

Performance of a Deep Learning Algorithm Compared with Radiologic Interpretation for Lung Cancer Detection on Chest Radiographs in a Health Screening Population

Background The performance of a deep learning algorithm for lung cancer detection on chest radiographs in a health screening population is unknown. Purpose To validate a commercially available deep learning algorithm for lung cancer detection on chest radiographs in a health screening population. Materials and Methods Out-of-sample testing of a deep learning algorithm was retrospectively performed using chest radiographs from individuals undergoing a comprehensive medical check-up between July 2008 and December 2008 (validation test). To evaluate the algorithm performance for visible lung cancer detection, the area under the receiver operating characteristic curve (AUC) and diagnostic measures, including sensitivity and false-positive rate (FPR), were calculated. The algorithm performance was compared with that of radiologists using the McNemar test and the Moskowitz method. Additionally, the deep learning algorithm was applied to a screening cohort undergoing chest radiography between January 2008 and December 2012, and its performances were calculated. Results In a validation test comprising 10 285 radiographs from 10 202 individuals (mean age, 54 years ± 11 [standard deviation]; 5857 men) with 10 radiographs of visible lung cancers, the algorithm's AUC was 0.99 (95% confidence interval: 0.97, 1), and it showed comparable sensitivity (90% [nine of 10 radiographs]) to that of the radiologists (60% [six of 10 radiographs]; P = .25) with a higher FPR (3.1% [319 of 10 275 radiographs] vs 0.3% [26 of 10 275 radiographs]; P < .001). In the screening cohort of 100 525 chest radiographs from 50 070 individuals (mean age, 53 years ± 11; 28 090 men) with 47 radiographs of visible lung cancers, the algorithm's AUC was 0.97 (95% confidence interval: 0.95, 0.99), and its sensitivity and FPR were 83% (39 of 47 radiographs) and 3% (2999 of 100 478 radiographs), respectively. Conclusion A deep learning algorithm detected lung cancers on chest radiographs with a performance comparable to that of radiologists, which will be helpful for radiologists in healthy populations with a low prevalence of lung cancer. © RSNA, 2020 Online supplemental material is available for this article. See also the editorial by Armato in this issue.

Role of chest radiographs in early lung cancer detection

Background

Lung cancer is the second most common and the most fatal form of cancer. Although annual low-dose computed tomography is used as the primary method of cancer screening, it presents challenges regarding resources as well as potential health risks from radiation exposure. Chest radiography (CXR), though less effective, is used frequently and commonly. Moreover, often in clinical settings, CXR is the first imaging modality used; computed tomography is subsequently performed if abnormalities are detected on CXRs. This study examined whether controlling for distractors and time constraints, as well as side-by-side comparison of multiple CXRs in clinical settings can aid earlier detection of radiological abnormalities indicative of lung cancer lesions.

Methods

Thirty-two attending physicians in the Republic of Korea examined 1,750 radiographs of 50 lung cancer cases. Using “hot spot” technology, participants indicated the possible locations of cancer lesions on each radiograph. Subsequently, the same radiographs, cropped to focus the anatomical regions where lung cancers were diagnosed, were shown side-by-side to the participants. The participants were asked to identify the radiograph which first enabled the diagnosis of lung cancer and which first showed a possible lesion.

Results

Removal of systemic constraints alone significantly improved lesion identification by 221.72±9.69 days. Presenting radiographs side-by-side, cropped to relevant areas, had an additional significant and positive impact on cancer detection in both hidden and open areas on CXRs. Also, lesions were detected at smaller sizes and earlier than when actually diagnosed.

Conclusions

CXR with improved methods and settings provides an easily accessible and low-risk imaging method for earlier detection of lung cancer compared to current clinical imaging settings. Further, this study demonstrates the potential effectiveness of programs that allow side-by-side comparisons of cropped areas of multiple radiographs to detect radiological abnormalities.

Chest radiography surveillance for lung cancer: Results from a National Health Insurance database in South Korea

BACKGROUND

Lung cancer screening with low-dose computed tomography reduced mortality in selected high risk patients. However, the use of chest radiography for lung cancer screening in Asian populations is still controversial. We investigated the effectiveness of chest radiographic surveillance using a nationwide health service data in South Korea.

METHODS

Data from the Korean National Health Insurance Service examinee cohort of 2004 to 2013 were examined, and 63,228 patients with lung cancer were identified, 38,494 (57%) of whom underwent chest radiography screening. The others did not undergo lung cancer screening and were considered as a control group. Clinical data including age, smoking, screening intervals, lung cancer stages, treatments, and survival were collected. Survival gain from surveillance after adjustment for lead-time bias based on the sojourn time was calculated. Cox-proportional hazard analyses were performed to evaluate the effectiveness of screening and to determine the appropriate screening interval for chest radiography surveillance.

RESULTS

Early lung cancer was found in 38% of patients receiving chest radiography versus 26% of those without surveillance. A patient age of more than 65 years (hazard ratio [HR], 1.53; 95% confidence limits [CL], 1.50-1.56), male (HR, 1.66; 95% CL, 1.62-1.70), and high lung cancer stages at the time of diagnosis were independent factors associated with mortality (each, P < 0.001). Chest radiography surveillance was a factor for decreasing mortality in female (HR, 0.81; 95% CL, 0.77-0.84, P < 0.001), with mortality reduction of 10% at the 3- and 5-year survival time-points. In female patients, chest radiography surveillance at intervals of less than 3 years was an independent predictor of improved survival.

CONCLUSIONS

Surveillance chest radiography increased survival in a female screened population in South Korea. Chest radiography intervals of less than 3 years may help to detect lung cancer in female patients.

Systematic review and critique of circulating miRNAs as biomarkers of stage I-II non-small cell lung cancer

Selected circulating microRNAs (miRNAs) have been suggested for non-invasive screening of non-small cell lung cancer (NSCLC), however the numerous proposed miRNA signatures are inconsistent. Aiming to identify miRNAs suitable specifically for stage I-II NSCLC screening in serum/plasma samples, we searched the databases "Pubmed", "Medline", "Scopus", "Embase" and "WOS" and systematically reviewed the publications reporting quantitative data on the efficacy [sensitivity, specificity and/or area under the curve (AUC)] of circulating miRNAs as biomarkers of NSCLC stage I and/or II. The 20 studies fulfilling the search criteria included 1110 NSCLC patients and 1009 controls, and were of medium quality according to Quality Assessment of Diagnostic Accuracy Studies checklist. In these studies, the patient cohorts as well as the control groups were heterogeneous for demographics and clinicopathological characteristics; moreover, numerous pre-analytical and analytical variables likely influenced miRNA determinations, and potential bias of hemolysis was often underestimated. We identified four circulating miRNAs scarcely influenced by hemolysis, each featuring high sensitivity (> 80%) and AUC (> 0.80) as biomarkers of stage I-II NSCLC: miR-223, miR-20a, miR-448 and miR-145; four other miRNAs showed high specificity (> 90%): miR-628-3p, miR-29c, miR-210 and miR-1244. In a model of two-step screening for stage I-II NSCLC using first the above panel of serum miRNAs with high sensitivity and high AUC, and subsequently the panel with high specificity, the estimated overall sensitivity is 91.6% and overall specificity is 93.4%. These and other circulating miRNAs suggested for stage I-II NSCLC screening require validation in multiple independent studies before they can be proposed for clinical application.

A comparison between quantitative PCR and droplet digital PCR technologies for circulating microRNA quantification in human lung cancer

Background

Selected microRNAs (miRNAs) that are abnormally expressed in the serum of patients with lung cancer have recently been proposed as biomarkers of this disease. The measurement of circulating miRNAs, however, requires a highly reliable quantification method. Quantitative real-time PCR (qPCR) is the most commonly used method, but it lacks reliable endogenous reference miRNAs for normalization of results in biofluids. When used in absolute quantification, it must rely on the use of external calibrators. Droplet digital PCR (ddPCR) is a recently introduced technology that overcomes the normalization issue and may facilitate miRNA measurement. Here we compared the performance of absolute qPCR and ddPCR techniques for quantifying selected miRNAs in the serum.

Results

In the first experiment, three miRNAs, proposed in the literature as lung cancer biomarkers (miR-21, miR-126 and let-7a), were analyzed in a set of 15 human serum samples. Four independent qPCR and four independent ddPCR amplifications were done on the same samples and used to estimate the precision and correlation of miRNA measurements obtained with the two techniques. The precision of the two methods was evaluated by calculating the Coefficient of Variation (CV) of the four independent measurements obtained with each technique. The CV was similar or smaller in ddPCR than in qPCR for all miRNAs tested, and was significantly smaller for let-7a (p = 0.028). Linear regression analysis of the miRNA values obtained with qPCR and ddPCR showed strong correlation (p < 0.001).

To validate the correlation obtained with the two techniques in the first experiment, in a second experiment the same miRNAs were measured in a larger cohort (70 human serum samples) by both qPCR and ddPCR. The correlation of miRNA analyses with the two methods was significant for all three miRNAs. Moreover, in our experiments the ddPCR technique had higher throughput than qPCR, at a similar cost-per-sample.

Conclusions

Analyses of serum miRNAs performed with qPCR and ddPCR were largely concordant. Both qPCR and ddPCR can reliably be used to quantify circulating miRNAs, however, ddPCR revealed similar or greater precision and higher throughput of analysis.

Electronic supplementary material

The online version of this article (doi:10.1186/s12896-016-0292-7) contains supplementary material, which is available to authorized users.

Lung cancer screening with low-dose computed tomography: Experiences from a tertiary hospital in Taiwan

BACKGROUND/PURPOSE

Lung cancer screening using low-dose computed tomography (CT) has been reported to reduce lung cancer-specific mortality for smokers at high risk. However, despite different characteristics of lung cancer in Asia, there are few data concerning this specific population for screening. We aim to analyze the performance of lung cancer screening with low-dose CT concurrent with chest radiography in Taiwan, with reference to international experience.

METHODS

During the 1-year period from January 2012 to December 2012, we conducted a retrospective, single-center population-based screening program for lung cancer in the setting of annual medical examinations. Participants were asymptomatic adults without prior history of any cancer. Low-dose CT and chest radiography were offered to all individuals. Baseline CT evaluations were defined as positive if any noncalcified nodule≥4 mm in diameter, which were then classified as solid, pure ground-glass or partial ground-glass opacity.

RESULTS

Of 3339 individuals, we detected 34 cancers, yielding an overall cancer detection rate of 1.02%. There was a particularly high cancer detection rate of 6.2% (8/129) in the high-risk group aged younger than 50 years with a positive family history of all types of cancers in first-degree relatives. Adenocarcinomas accounted for 88% (30/34) of cancers and 99% of them were early-stage (including carcinoma in situ and Stage I). The probability of cancers was significant higher in nodules with interval growth (odds ratio 257.89, p = 0.0002). There was no significant difference in the probability of cancers between ground glass opacity nodules and solid nodules (odds ratio 1.16, p=0.72). Of all screen-detected cancers, 61.76% (21/34) were chest radiographically occult.

CONCLUSION

Low-dose CT is effective to detect early lung cancers. Further establishment of selection criteria for lung cancer screening, specifically for Asian individuals, is definitely warranted.

Could cigarette packaging be used as a tool to make prevention of smoke-induced respiratory diseases?

The most important consequences of smoking are chronic obstructive pulmonary disease (COPD) and lung cancer (LC). Although the use of shocking images and warning messages on cigarette packaging is a valid tool of smoke dishabituation, unfortunately, millions of people go on smoking. Our hypotheses is that cigarette packet covers could also be used to give further messages, especially meant to spur also a screening of smoke-induced respiratory diseases. Messages on cigarette packaging suggesting smokers to perform a spirometry and a chest X-ray may persuade them not only to quit their habit but also to have a screening for COPD and LC prevention. If our hypotheses is taken into account it will have a strong worldwide impact.

Resection rate of lung cancer in Teesside (UK) and Varese (Italy): a comparison after implementation of the National Cancer Plan

BACKGROUND

In a lung cancer survey in 2000 we showed significantly less favourable stage distribution and lower resection rate in Teesside (UK) than in the comparable industrialised area of Varese (Italy). Lung cancer services in Teesside were subsequently reorganised according to National Cancer Plan recommendations.

METHODS

For all new lung cancer cases diagnosed in Teesside (n=324) and Varese (n=260) during the 12 months October 2010 to September 2011 (hereafter 'the 2010 cohort'), demographic, clinico-pathological and disease management data were prospectively recorded using the same database and protocol as the 2000 survey. Findings were analysed focusing on resection rate.

RESULTS

In the 2010 cohort compared with 2000, both in Teesside and Varese emergency referral decreased (p<0.001), performance status improved (p<0.001), but cancer stage shift was not seen; resection rate improved in Teesside, from 7% to 11% (p=0.054), and was unchanged in Varese (24%). Moreover, in Teesside compared with Varese the stage distribution remained less favourable, stage I-II non-small cell lung cancer (NSCLC) proportion being respectively 12% and 19% (p=0.040), and resection rate in all lung cancers remained lower (11% and 24%; p<0.001). On multivariate analysis, resection predictors in Teesside were as follows: stage I-II NSCLC (OR 86.14; 95% CI 31.80 to 233.37), performance status 0-1 (OR 5.02; 95% CI 1.48 to 17.07), belonging to 2010 cohort (OR 2.85; 95% CI 1.06 to 7.64).

CONCLUSIONS

In Teesside the main independent predictor of resection was disease stage; in 2010-2011 compared with 2000, lung cancer service improved but stage shift did not occur, and resection rate increased but remained significantly lower than in Varese.

Is chest X-ray screening for lung cancer in smokers cost-effective? Evidence from a population-based study in Italy

Background

After implementation of the PREDICA annual chest X-ray (CXR) screening program in smokers in the general practice setting of Varese-Italy a significant reduction in lung cancer-specific mortality (18 %) was observed. The objective of this study covering July 1997 through December 2006 was to estimate the cost-effectiveness of this intervention.

Methods

We examined detailed information on lung cancer (LC) cases that occurred among smokers invited to be screened in the PREDICA study (Invitation-to-screening Group, n = 5815 subjects) to estimate costs and quality-adjusted life-years (QALYs) from LC diagnosis until death. The control group consisted of 156 screening-eligible smokers from the same area, uninvited and unscreened, who developed LC and were treated by usual care. We calculated the incremental net monetary benefit (INMB) by comparing LC management in screening participants (n = 1244 subjects) and in the Invitation-to-screening group versus control group.

Results

The average number of QALYs since LC diagnosis was 1.7, 1.49 and 1.07, respectively, in screening participants, the invitation-to-screening group, and the control group. The average total cost (screening + management) per LC case was higher in screening participants (€17,516) and the Invitation-to-screening Group (€16,167) than in the control group (€15,503). Assuming a maximum willingness to pay of €30,000/QALY, we found that the intervention was cost-effective with high probability: 79 % for screening participation (screening participants vs. control group) and 95 % for invitation-to-screening (invitation-to-screening group vs. control group).

Conclusions

Based on the PREDICA study, annual CXR screening of high-risk smokers in a general practice setting has high probability of being cost-effective with a maximum willingness to pay of €30,000/QALY.

Impact of endoscopic screening on mortality reduction from gastric cancer

AIM: To investigate mortality reduction from gastric cancer based on the results of endoscopic screening.

METHODS: The study population consisted of participants of gastric cancer screening by endoscopy, regular radiography, and photofluorography at Niigata city in 2005. The observed numbers of cumulative deaths from gastric cancers and other cancers were accumulated by linkage with the Niigata Prefectural Cancer Registry. The standardized mortality ratio (SMR) of gastric cancer and other cancer deaths in each screening group was calculated by applying the mortality rate of the reference population.

RESULTS: Based on the results calculated from the mortality rate of the population of Niigata city, the SMRs of gastric cancer death were 0.43 (95%CI: 0.30-0.57) for the endoscopic screening group, 0.68 (95%CI: 0.55-0.79) for the regular radiographic screening group, and 0.85 (95%CI: 0.71-0.94) for the photofluorography screening group. The mortality reduction from gastric cancer was higher in the endoscopic screening group than in the regular radiographic screening group despite the nearly equal mortality rates of all cancers except gastric cancer.

CONCLUSION: The 57% mortality reduction from gastric cancer might indicate the effectiveness of endoscopic screening for gastric cancer. Further studies and prudent interpretation of results are needed.

[The six steps necessary in elaborating critically appraised topics]

Different types of critically appraised topics (CATs) can be elaborated in diagnostic imaging: comparison of diagnostic tests, evaluation of techniques for early detection (screening), economical analyses, or therapeutic aspects, among others. Their design will vary in function of the question they aim to answer. For example, for treatment evaluation, clinical trials are the best, but if there are secondary studies (systematic reviews or meta-analyses) that synthesize information from several studies, the results will be more important and the scientific conclusions will be more relevant. Regardless of the study design used, the elaboration of a CAT will involve six steps: 1) question; 2) systematic and efficient bibliographic search; 3) levels of evidence (choosing the articles that have the best level); 4) critical reading of the articles chosen; 5) applying conclusions to the context, and 6) recommendations. In this article, we will describe these steps and the nuances for different types of studies in each step.

Lung cancer screening-don’t forget the chest radiograph

Lung cancer is a major health burden and early detection only bears the possibility of curative treatment. Screening with computed tomography (CT) recently demonstrated a mortality reduction in selected patients and has been incorporated in clinical guidelines. Problems of screening with CT are the excessive number of false positive findings, costs, radiation burden and from a global point of view shortage of CT capacity. In contrast, chest radiography could be an ideal screening tool in the early detection of lung cancer. It is widely available, easy to perform, cheap, the radiation burden is negligible and there is only a low rate of false positive findings. Large randomized controlled trials could not show a mortality reduction, but different large population-based cohort studies have shown a lung cancer mortality reduction. It has been argued that community-based cohort studies are more closely reflecting the “real world” of everyday medicine. Radiologists should be aware of the found mortality reduction and realize that early detection of lung cancer is possible when reading their daily chest radiographs. Offering a chest radiograph in selected scenarios for the early detection of lung cancer is therefore still justified.

Chest X-ray screening for lung cancer: overdiagnosis, endpoints, and randomized population trials

Publication of the National Lung Screening Trial (NLST) generated excitement by concluding that CT screening reduces lung cancer mortality when compared to chest X-ray (CXR) screening. In contrast, CXR screening has long been considered to be ineffective. This is because randomized population trials (RPTs) have failed to demonstrate significant mortality reductions in populations randomized to CXR screening. While these studies demonstrate that CXR screening is associated with significant survival advantages, these advantages have been widely interpreted as spurious, due to the inference that CXR screening leads to substantial lung cancer overdiagnosis. Indeed, the reality of the overdiagnosis hypothesis is the only alternative to the conclusion that CXR screening was effective in these trials and that survival more accurately reflected the benefit of CXR screening than mortality. Mortality comparisons would be biased if randomization fails to create comparison groups with an equal probability of mortality from the target cancer. The objective of this manuscript is to review existing RPTs on CXR screening for lung cancer, and to analyze which endpoint most accurately reflects screening efficacy. We conclude that the evidence supports that CXR screening is superior to no screening, and the magnitude of overdiagnosis is minimal in the context of CXR screening.