Original Research
Cardiothoracic Imaging
November 10, 2021

Cancer Risk in Nodules Detected at Follow-Up Lung Cancer Screening CT

Abstract

BACKGROUND. Nodules may have different lung cancer risks when new on follow-up CT versus when present on previously performed CT (i.e., existing nodules). Diameter-based Lung-RADS and volume-based NELSON (Nederlands-Leuvens Longkanker Screenings ONderzoek trial) categories have shown variable performance in nodule risk assessment.
OBJECTIVE. The purpose of this study was to assess Lung-RADS and NELSON classifications of nodules detected on follow-up lung cancer screening CT examinations.
METHODS. This retrospective study included 185 patients (100 women and 85 men; median age, 66 years) who underwent a lung cancer screening CT examination for which a prior CT examination was available. Stratified random sampling was performed to enrich the sample with suspicious nodules, yielding 50, 45, 47, 30, and 13 nodules with Lung-RADS categories 2, 3, 4A, 4B, and 4X, respectively. Lung-RADS categories were recorded from clinical reports. The linear measurements of the nodules were extracted from clinical reports to generate Lung-RADS categories by use of strict criteria from Lung-RADS version 1.1. Two radiologists used a semiautomated tool to obtain nodule volumes, which were used to generate NELSON categories. Lung cancer risk was assessed. ROC analysis was performed. Percentages and AUCs were weighted on the basis of Lung-RADS category frequencies in the underlying screening cohort.
RESULTS. Twenty-nine cancers were diagnosed. The weighted cancer risk was 5% for new nodules, 1% for stable existing nodules, and 44% for growing existing nodules. None of the clinical Lung-RADS category 2 nodules were cancer. With use of strict Lung-RADS version 1.1 criteria, 34 nodules, including seven cancers, were downgraded to category 2. The AUC for cancer was 0.96 for clinical Lung-RADS, 0.81 for strict Lung-RADS, 0.71–0.84 for the NELSON algorithm (two readers), and 0.89 for nodule diameter measurement. Clinical Lung-RADS achieved weighted sensitivity and specificity, respectively, of 100% and 85% for the entire sample, 100% and 41% for new nodules, and 100% and 94% for existing nodules. The optimal diameter threshold was 8 mm for existing nodules versus 6 mm for new nodules.
CONCLUSION. Lung-RADS, as applied by radiologists in clinical practice, achieved excellent performance on follow-up screening examinations. Strict Lung-RADS resulted in the downgrading of some cancers to category 2. Volumetric assessments had weaker performance than clinical Lung-RADS. New nodules warrant smaller size thresholds than existing nodules.
CLINICAL IMPACT. The findings of the present study provide insight into radiologists' management of nodules detected on follow-up screening examinations.

HIGHLIGHTS

Key Finding
Weighted cancer risk on follow-up screening examinations was 5% for new nodules, 1% for stable existing nodules, and 44% for growing existing nodules. The weighted AUC was 0.96 for clinical Lung-RADS categories versus 0.71–0.84 for volumetric NELSON categories. Application of strict Lung-RADS criteria downgraded seven of 29 cancers to Lung-RADS category 2.
Importance
Lung-RADS, as applied by radiologists in clinical practice, had excellent performance for cancer risk assessment on follow-up screening examinations. Strict Lung-RADS criteria downgraded some malignancies.
With increasing utilization of CT for lung cancer screening examinations, algorithms to triage pulmonary nodules are critically important to provide optimal management of the patients who undergo such screening. Various approaches have been described to guide the evaluation of nodules detected in lung cancer screening. These include the American College of Radiology Lung-RADS [1], which is primarily based on linear measurement of nodule size, as well as the algorithm used in the Dutch NELSON (Nederlands-Leuvens Longkanker Screenings ONderzoek) trial [2], which is primarily based on nodule volume and growth rate.
The approach to nodules identified on follow-up lung cancer screening CT is somewhat different from the approach to nodules seen on baseline CT. Nodules that are new on follow-up CT may have a higher risk of lung cancer than nodules on follow-up CT that were also present on baseline CT (i.e., existing nodules) [3]. Certain nodule triage algorithms, such as those from the Fleischner Society [4] and the original NELSON trial, use the same size cutoffs for new nodules on follow-up CT as for nodules on baseline CT. Conversely, Lung-RADS recommends a lower size threshold for raising suspicion for cancer for new nodules on follow-up CT than for nodules on baseline CT. For existing nodules at follow-up, both Lung-RADS and the NELSON algorithm use growth since the earlier examination for determining the need for follow-up: nodules that either are not growing or (for the NELSON algorithm) are slowly growing and do not warrant close follow-up.
A number of studies have compared the performance of nodule risk assessment algorithms (e.g., the National Lung Screening Trial and NELSON trial cohorts [57]) and have yielded inconsistent findings regarding the relative performance of diameter-based Lung-RADS versus the volume-based NELSON algorithm. However, to our knowledge, none of these studies specifically focused on nodules detected on follow-up screening CT examinations. Therefore, in the present study, we evaluated the utility of both risk assessment algorithms (Lung-RADS and NELSON) for nodules detected on follow-up lung cancer screening CT examinations.

Methods

This retrospective HIPAA-compliant study was approved by the institutional review board with a waiver of informed consent.

Patient Selection

We performed an automated search of a database of 5835 patients who underwent lung cancer screening CT within our health care network (which included two academic sites and one community site) between July 2015 and August 2018, to identify those patients who also had a prior chest CT examination within our network. This search yielded a total of 1150 patients who underwent a total of 2242 CT examinations with a prior examination. On the basis of the original clinical interpretations, 513 of these examinations were classified as Lung-RADS category 1; of the remainder, 1435 (83%) were classified as category 2, 166 (10%) as category 3, 80 (5%) as category 4A, 34 (2%) as category 4B, and 14 (1%) as category 4X. For each Lung-RADS category (except for categories 1, 4B, and 4X), we randomly selected examinations, with a prior specified maximum of 50 examinations per category (aside from category 1, for which no examinations were selected). For categories 4B and 4X, all examinations were initially selected given that fewer than 50 examinations were available in each category. This initial random selection process yielded 198 examinations for 185 patients. After the initial selection, it was discovered that two Lung-RADS category 2 nodules (per the clinical reports) were misclassified by the search algorithm (one as category 4A and the other as category 4B); these two nodules were reclassified as category 2 on the basis of the clinical reports. After this reclassification, an additional category 4B nodule was randomly selected to replace the reclassified category 4B nodule. Finally, the first follow-up examination was selected for any of the 185 patients who had multiple follow-up examinations. Following this process, the study sample included 185 follow-up screening examinations of 185 patients (100 women and 85 men; median age, 66 years) in whom a nodule was detected (50 Lung-RADS category 2 nodules, 45 category 3 nodules, 47 category 4A nodules, 30 category 4B nodules, and 13 category 4X nodules). Patient demographics and family history of lung cancer were extracted from the electronic medical records. For examinations that detected multiple nodules, only the nodule determining the overall category assigned in the report was considered in the analysis.

Clinical Lung-RADS and Strict Lung-RADS Categories

The examinations had been interpreted using Lung-RADS version 1.0, given the period in which the study was conducted. A total of 181 examinations were interpreted by fellowship-trained cardiothoracic radiologists at one of the two academic sites; four examinations were interpreted by radiologists at the community site. The Lung-RADS categories were extracted automatically from the clinical reports and then were corrected if they did not match the clinical report on further manual review. Reports were also manually reviewed to extract nodule attenuation (classified as solid, part solid, or ground-glass), nodule size on the present CT examination, and nodule size on the prior CT examination. The clinically recorded nodule size measurements typically represented the mean of the long- and short-axis diameters of the nodules, per institutional practice. If the current examination did not report the nodule size from the prior examination, then the size measurement was obtained from the prior report. These size measurements were then used to retrospectively determine a Lung-RADS category for each nodule, on the basis of the size thresholds for new and existing nodules in Lung-RADS version 1.1 (strict Lung-RADS). Clinical Lung-RADS category 4X was not adjusted because this category depends on additional aspects of nodule appearance, and the images were not reviewed for the purposes of this assessment.

Retrospective Determination of Nodule Volumes and NELSON Categories

Two fellowship-trained thoracic radiologists (M.M.H. and S.C.B., both with 6 years of posttraining experience) independently reviewed the images of each patient. The radiologists accessed the clinical reports to identify the nodule on which the category determination was based, but they were blinded to subsequent imaging and clinical diagnoses. To derive NELSON categories for each nodule, the radiologists classified the nodule as new or existing, assessed its attenuation (solid, part solid, or ground-glass), and determined its volume using a semiautomated segmentation software (syngo.via, version VB40, Siemens Healthcare). For nodules classified as existing at the time of this retrospective review, the radiologists segmented the nodule on both the prior and current examinations. For part-solid nodules, the radiologist separately segmented both the total nodule and the solid component of the nodule. By use of the readers' nodule assessments and volu-metric measurements, NELSON categories were assigned for each nodule for each reader [2]. The NELSON algorithm categorizes new nodules (as well as nodules observed on baseline examinations) as NODCAT 1, 2, 3, or 4, depending on nodule attenuation and volume. The NELSON algorithm categorizes existing nodules as GROWCAT A, B, or C, depending on volume doubling times (> 600 days, 400–600 days, or 0–399 days, respectively).

Reference Standard

Electronic medical records were manually reviewed through March 2021 to identify any clinical diagnoses of lung cancer attributed to the nodule on which the category determination was based for each patient. For patients without histologic confirmation, empirical treatment with radiation therapy was considered to indicate an empirical diagnosis of lung cancer.

Data Analysis

The characteristics of the patients and nodules were described using summary characteristics. Characteristics were compared between new and existing nodules by use of the Wilcoxon and Fisher exact tests. Results were weighted according to the relative frequencies of the Lung-RADS categories in the entire lung cancer screening population of 1150 patients (with 2242 CT examinations) from which the study sample was drawn. This weighting was intended to provide representative cancer risks for an overall screening population (e.g., a population in which the Lung-RADS category 2 is much more common than Lung-RADS categories 3 and 4), given the selection of similar numbers of nodules for each category in the present study sample. The weighting of categories was not performed when the cancer frequency of individual clinical Lung-RADS categories was summarized. The frequency of lung cancer was summarized, as stratified by combinations of clinical Lung-RADS versus strict Lung-RADS, new versus existing nodules, and Lung-RADS category. By use of the Fisher exact test, these frequencies were compared between new and existing nodules for each clinical Lung-RADS category. The weighted cancer frequency was summarized for new nodules, stable existing nodules, and growing existing nodules and was compared using the Fisher exact test. Characteristics of cancers on follow-up CT examinations were summarized, stratified by new nodules, stable existing nodules, and growing existing nodules. Nodules that were reclassified when strict Lung-RADS was used were summarized in terms of the reassignments, the reasons for the reassignments, and the characteristics of the cancers among the reassigned nodules. Interreader agreement for nodule volumes was assessed using the intra-class correlation coefficient. Lung cancer risk was also summarized by NELSON category, stratified by reader and new versus existing nodules. Sensitivity and specificity for lung cancer were assessed for clinical Lung-RADS, strict Lung-RADS, the NELSON algorithm, and nodule diameter measurement, for all nodules as well as separately for new and existing nodules; to allow comparison of approaches for the same sets of nodules, the sensitivity and specificity of the NELSON algorithm were computed on the basis of the designation of nodules as new or existing in the clinical reports. The AUC for lung cancer according to the various approaches was also computed for all nodules but was not calculated separately for new and existing nodules given the small number of cancers among new nodules. All AUCs were calculated using the previously described weighting method and 95% CIs were generated by bootstrapping with 1500 samples. Clinical Lung-RADS and strict Lung-RADS were defined as positive at category 3 or greater. The NELSON algorithm was defined as positive at NODCAT 3 or greater for new nodules or at GROWCAT B or greater for existing nodules. For nodule diameter measurement, the optimal cutoff was identified using the Youden index from ROC analysis. Differences in proportions were evaluated using the Fisher exact test. A p value of less than .05 was considered statistically significant. Data were initially entered in a secure Web application for research electronic data capture (REDCap, version 11, REDCap Consortium) [8] and then were downloaded and analyzed with predictive analytics software (JMP Pro, version 15.2, SAS Institute).

Results

Patients and Nodules

Characteristics of the 185 patients and nodules are presented in Table 1. The median interval between the prior and current CT examinations was 12 months (interquartile range, 8–15 months; range, 1–96 months). The median nodule diameter was 7 mm (range, 2–30 mm). Of the 185 nodules, 140 (76%) were solid, 30 (16%) were part solid, and 15 (8%) were ground-glass. A total of 104 nodules (56%) were existing, whereas 81 (44%) were new. New and existing nodules were not significantly different in terms of patient age, patient sex, nodule morphology, or nodule size (all p > .05). A total of 29 nodules (16%) were diagnosed as lung cancer (24 histologically and five empirically, given treatment with radiation therapy). These malignant nodules had a median diameter of 10 mm (range, 6–25 mm).
TABLE 1: Characteristics of Patients and Nodules
CharacteristicAll Nodules (n = 185)New Nodules (n = 81)Existing Nodules (n = 104)pa
Age (y), median (range)66 (55–79)65 (56–79)67 (55–78).35
Sex    
 Female100 (54)44 (54)56 (54).53
 Male85 (46)37 (46)48 (46) 
Nodule attenuation   .44
 Solid140 (76)65 (80)75 (72) 
 Part-solid30 (16)10 (12)20 (19) 
 Ground-glass15 (8)6 (7)9 (9) 
Nodule diameter (mm), median (range)7 (2–30)7 (2–29)7 (2–30).78
Clinical Lung-RADS category   < .001
 250 (27)4 (5)46 (44) 
 345 (24)30 (37)15 (14) 
 4A47 (25)27 (33)20 (19) 
 4B30 (16)18 (22)12 (12) 
 4X13 (7)2 (2)11 (11) 
Lung cancerb29 (3)5 (5)24 (2).60

Note—Except where otherwise indicated, data are number of patients, with percentage of patients in parentheses.

a
New versus existing nodules.
b
Weighted percentages are based on relative frequencies of clinical Lung-RADS categories in entire lung cancer screening population from which study sample was drawn.

Risk of Lung Cancer by Lung-RADS Category

The risk of lung cancer in new and existing nodules, as stratified by Lung-RADS category, is presented in Table 2. Based on clinical Lung-RADS, the frequency of lung cancer in new and existing nodules was 0% (0/4) and 0% (0/46) in category 2 nodules (p > .99), 10% (3/30) and 7% (1/15) in category 3 nodules (p = .59), 4% (1/27) and 30% (6/20) in category 4A nodules (p = .02), 0% (0/19) and 67% (8/12) in category 4B nodules (p < .001), and 50% (1/2) and 82% (9/11) in category 4X nodules (p = .42).
TABLE 2: Risk of Lung Cancer, as Stratified by Lung-RADS Category
CategoryClinical Lung-RADSStrict Lung-RADSa
All NodulesNew NodulesExisting NodulesAll NodulesNew NodulesExisting Nodules
20/50 (0)0/4 (0)0/46 (0)7/83 (1)1/10 (3)7/73 (1)
34/45 (9)3/30 (10)1/15 (7)0/18 (0)0/18 (0)0/0 (0)
4A7/47 (15)1/27 (4)6/20 (30)2/27 (9)2/24 (10)0/3 (0)
4B8/30 (27)0/18 (0)8/12 (67)10/44 (22)1/27 (10)9/17 (39)
4X10/13 (77)1/2 (50)9/11 (82)10/13 (77)1/2 (50)9/11 (82)

Note—Except where otherwise indicated, data are numerators and denominators, with percentage in parentheses.

a
Percentages are weighted by relative frequency of clinical Lung-RADS scores in the underlying population.
The frequency of lung cancer was as follows: five of 81 new nodules, 10 of 79 stable existing nodules, and 14 of 25 growing existing nodules. These frequencies corresponded with weighted cancer rates of 5% for new nodules, 1% for stable existing nodules, and 44% for growing existing nodules. Overall, the frequency of cancer was not significantly different between new nodules and stable existing nodules (p = .14) or between growing nodules and new nodules (p = .05); cancer frequency was significantly higher in growing nodules versus stable existing nodules (p = .003). Table 3 provides a comparison of patient age, patient sex, nodule attenuation, nodule diameter, and both clinical and strict Lung-RADS categories for the cancers in new, stable, and growing nodules on follow-up CT.
TABLE 3: Characteristics of 29 Cancers on Follow-Up CT Examinations
CharacteristicAll Nodules (n = 29)New Nodules (n = 5)Stable Existing Nodules (n = 10)Growing Existing Nodules (n = 14)
Age (y). median (range)66 (55–78)66 (58–74)71 (63–78)63 (55–75)
Sex    
 Female17 (59)2 (40)7 (70)8 (57)
 Male12 (41)3 (60)3 (30)6 (43)
Nodule attenuation    
 Solid13 (45)3 (60)4 (40)6 (43)
 Part-solid13 (45)1 (20)6 (60)6 (43)
 Ground-glass3 (10)1 (20)0 (0)2 (14)
Nodule diameter (mm), median (range)10 (6–25)10 (6–19)10 (8–25)10 (6–21)
Clinical Lung-RADS category    
 20 (0)0 (0)0 (0)0 (0)
 34 (14)3 (60)1 (10)0 (0)
 4A7 (24)1 (20)4 (40)2 (14)
 4B8 (28)0 (0)1 (10)7 (50)
 4X10 (34)1 (20)4 (40)5 (36)
Strict Lung-RADS category    
 27 (24)1 (20)6 (60)0 (0)
 30 (0)0 (0)0 (0)0 (0)
 4A2 (7)2 (20)0 (0)0 (0)
 4B10 (34)1 (20)0 (0)9 (64)
 4X10 (34)1 (20)4 (40)5 (35)

Note—Except where otherwise indicated, data are number of patients, with percentage in parentheses.

Strict Lung-RADS resulted in a change in Lung-RADS category for 59 of the 185 nodules:. One category 2 nodule was changed to category 3. Of the 29 category 3 nodules, 18 were changed to category 2, seven were changed to category 4A, and four were changed to category 4B. Of the 27 category 4A nodules, 14 were changed to category 2, one was changed to category 3, and 12 were changed to category 4B. The two category 4B nodules were changed to category 2. A total of 34 nodules were downgraded from clinical Lung-RADS categories of 3, 4A, or 4B to strict Lung-RADS category 2, for the following reasons: unchanged size since the prior examination (i.e., did not exhibit an increase in size of > 1.5 mm) (n = 23); a ground-glass nodule measuring less than 30 mm (n = 10); and a new solid nodule measuring less than 4 mm (n = 1). As a net result of all adjustments in Lung-RADS categories, 83 nodules were category 2 by use of strict Lung-RADS, compared with 50 that were category 2 by use of clinical Lung-RADS. None of the 52 clinical Lung-RADS category 2 nodules were cancer. However, the nodules reclassified as category 2 using strict Lung-RADS included seven cancers, representing 24% of the 29 cancers in the study cohort. These seven cancers that were reclassified as Lung-RADS category 2 included six nodules that were unchanged in size (Fig. 1) and one existing ground-glass nodule measuring less than 30 mm (Fig. 2).
Fig. 1A —77-year-old woman with nodule in left lower lobe.
A, Current lung cancer screening CT examination (A) and lung cancer CT performed 3 months earlier (B) show what was described as stable solid nodule with mean diameter of 8 mm. Nodule was categorized as Lung-RADS 4A in clinical report but was described as Lung-RADS 2 by strict application of size criteria. Subsequent biopsy showed adenocarcinoma.
Fig. 1B —77-year-old woman with nodule in left lower lobe.
B, Current lung cancer screening CT examination (A) and lung cancer CT performed 3 months earlier (B) show what was described as stable solid nodule with mean diameter of 8 mm. Nodule was categorized as Lung-RADS 4A in clinical report but was described as Lung-RADS 2 by strict application of size criteria. Subsequent biopsy showed adenocarcinoma.
Fig. 2A —58-year-old woman with nodule in right lower lobe.
A, Current lung cancer screening CT examination (A) and lung cancer screening CT examination performed 2 years earlier (B) show nodule that was described as growing pure ground-glass nodule (associated with cystic lesion) with mean diameter of 15 mm. Nodule was categorized as Lung-RADS 3 in clinical report but was described as Lung-RADS 2 by strict application of size criteria. Subsequent wedge resection showed lepidic-predominant adenocarcinoma.
Fig. 2B —58-year-old woman with nodule in right lower lobe.
B, Current lung cancer screening CT examination (A) and lung cancer screening CT examination performed 2 years earlier (B) show nodule that was described as growing pure ground-glass nodule (associated with cystic lesion) with mean diameter of 15 mm. Nodule was categorized as Lung-RADS 3 in clinical report but was described as Lung-RADS 2 by strict application of size criteria. Subsequent wedge resection showed lepidic-predominant adenocarcinoma.

Volumetric Analysis Using the NELSON Algorithm

Interreader agreement for nodule volume, expressed as the intraclass correlation coefficient, was 0.98 (95% CI, 0.97–0.99). The frequency of lung cancer according to NELSON categories is presented in Table 4. The weighted rates of lung cancer for new nodules were as follows: 0% for NODCAT 2 for both readers; 3% and 6% for NODCAT 3 for readers 1 and 2, respectively; and 5% for NODCAT 4 for both readers. The weighted rates of lung cancer for existing nodules were as follows: 1% for GROWCAT A for both readers; 5% and 9% for GROWCAT B for readers 1 and 2, respectively; and 10% and 64% for GROWCAT C for readers 1 and 2, respectively.
TABLE 4: Risk of Lung Cancer by NELSON Category
NELSON CategoryReader 1Reader 2
New  
 NODCAT20/10 (0)0/13 (0)
 NODCAT31/46 (3)3/48 (6)
 NODCAT 41/17 (5)1/18 (5)
Existinga  
 GROWCAT A10/71 (1)8/78 (1)
 GROWCAT B5/10 (5)2/6 (9)
 GROWCAT C12/31 (10)15/22 (64)

Note—Data are numerators and denominators, with percentage or weighted percentage in parentheses. NELSON = Nederlands-Leuvens Longkanker Screenings ONderzoek trial, NODCAT = NELSON categorization of new nodules, GROWCAT = NELSON categorization of existing nodules.

a
Percentages are weighted according to relative frequency of clinical Lung-RADS scores in the underlying population. Growth categories correspond to volume doubling times of more than 600 days (category A), 400–600 days (category B), or 0–399 days (category C).

Comparison of Methods for Lung Cancer Detection

The diagnostic performance of clinical Lung-RADS, strict Lung-RADS, the NELSON algorithm, and nodule diameter measurements for lung cancer detection is shown in Table 5. For all nodules observed on follow-up CT, the AUC was 0.96 for clinical Lung-RADS, 0.81 for strict Lung-RADS, 0.71–0.84 for NELSON for the two readers, and 0.89 for nodule diameter measurement. The optimal threshold for the nodule diameter was 8 mm. Sensitivity and specificity, respectively, were as follows: 100% and 85% for clinical Lung-RADS, 68% and 88% for strict Lung-RADS, 64–75% and 75–88% for the NELSON algorithm, and 86% and 84% for nodule diameter measurement.
TABLE 5: Diagnostic Performance of Nodule Evaluation Approaches
ApproachAUC (95% CI)All NodulesaNew NodulesaExisting Nodulesa
SensitivitySpecificitySensitivitySpecificitySensitivitySpecificity
Clinical Lung-RADS0.96 (0.93–0.98)29/29 (100)50/156 (85)5/5 (100)4/76 (41)24/24 (100)46/80 (94)
Strict Lung-RADSb0.81 (0.68–0.90)22/29 (68)76/156 (88)4/5 (74)9/76 (38)18/24 (65)67/80 (98)
NELSONc       
 Reader 10.71 (0.60–0.83)19/29 (64)71/156 (75)3/5 (48)16/76 (46)16/24 (71)55/80 (81)
 Reader 20.84 (0.73–0.92)21/29 (75)100/156 (88)5/5 (100)16/76 (46)16/24 (65)67/80 (96)
Nodule diameter measurementb0.89 (0.82–0.95)26/29 (86)100/156 (84)5/5 (100)25/76 (62)23/24 (96)56/80 (85)

Note—Except where otherwise indicated, data are numerators and denominators, with percentage or weighted percentage in parentheses. NELSON = Neder-lands-Leuvens Longkanker Screenings ONderzoek trial.

a
Weighted based on relative frequencies of clinical Lung-RADS categories in entire lung cancer screening population from which study sample was drawn.
b
Considered positive at category of 3 or higher.
c
Considered positive at a NELSON new nodule categorization of NODCAT 2 or higher or a NELSON existing nodule categorization of GROWCAT B or higher.
d
Based on Youden index from ROC analysis and considered positive at 8 mm for all nodules, 6 mm for new nodules, and 8 mm for existing nodules.
For new nodules, the optimal threshold for the nodule diameter was 6 mm. Sensitivity and specificity, respectively, were 100% and 41% for clinical Lung-RADS, 74% and 38% for strict Lung-RADS, 48–100% and 46% for the NELSON algorithm, and 100% and 62% for nodule diameter measurement. For existing nodules, the optimal threshold for the nodule diameter was 8 mm. Sensitivity and specificity, respectively, were 100% and 94% for clinical Lung-RADS, 65% and 98% for strict Lung-RADS, 65–71% and 81–96% for the NELSON algorithm, and 96% and 85% for nodule diameter measurement. The use of a diameter threshold of 8 mm for new nodules resulted in sensitivity of 61% (3/5) and specificity of 82% (44/76). The use of a threshold of 6 mm for existing nodules resulted in a sensitivity of 100% (24/24) and specificity of 64% (35/80).

Discussion

We evaluated the performance of the Lung-RADS and NELSON nodule classification schemes for nodules detected at follow-up lung cancer screening CT. The frequency of cancer was 1% in stable existing nodules, 5% in new nodules, and 44% in growing existing nodules. Among both new nodules and existing nodules, the frequency of cancer increased with increasing Lung-RADS categories. Clinical Lung-RADS showed excellent sensitivity and specificity for detecting cancer in existing nodules and excellent sensitivity for detecting cancer in new nodules, although it showed low specificity in new nodules. Volumetric assessment using the NELSON scheme had a lower diagnostic performance than clinical Lung-RADS scores. A nodule diameter cutoff of 8 mm yielded 96% sensitivity in existing nodules but only 61% in new nodules; rather, a smaller cutoff of 6 mm was optimal for new nodules. The lower size threshold for new versus existing nodules is consistent with the smaller thresholds for new nodules in the Lung-RADS recommendations [1].
Strict application of the Lung-RADS criteria (i.e., adjusting nodule categories on the basis of size and growth criteria, aside from category 4X which was not adjusted) resulted in the downgrading of many nodules from categories 3 through 4B to category 2, most commonly because they were stable nodules or were pure ground-glass nodules measuring less than 30 mm. Although clinical Lung-RADS categories had 100% sensitivity for cancer at a threshold category of 3 or higher, the nodules that were downgraded to category 2 included numerous cancers. Downgrading of malignancies based on strict application of Lung-RADS criteria indicates that the interpreting radiologists had exercised discretion beyond size measurements in the usage of Lung-RADS recommendations, so as to report higher categories for nodules that they believed were likely malignant. In other words, radiologists likely strongly considered additional nodule features beyond size measurements when categorizing nodules. Current practices for assessing stable nodules that do not show entirely benign characteristics (including characteristics other than spiculation) are inconsistent because radiologists may variably elect to use categories lower than category 4X to describe nodules with suspicious features if the nodules are stable or show minimal growth. Our results support category 4X as the most appropriate category for nodules that do not exhibit entirely benign characteristics, even if stable. The use of category 4X in this context is consistent with descriptions in the Lung-RADS document.
Although use of Lung-RADS category 4X in the described fashion helps maximize cancer detection, whether maximal sensitivity should be sought when using Lung-RADS is unclear. Pure ground-glass nodules show indolent behavior, with a very low risk of metastatic disease or recurrence after resection, and follow-up rather than immediate therapy of such nodules is supported by cost-effectiveness analysis [9]. The management approach for slowly growing solid nodules is controversial. Surgeons may prefer early definitive therapy with surgical resection of solid nodules that are suspicious for lung cancer. However, the NELSON algorithm views growth rate as a central determinant of when to intervene for lung nodules, including those that are solid [2, 10].
The volumetric NELSON algorithm, which is based on nodule growth, showed poorer performance than the clinical Lung-RADS categories. This likely reflects the presence of slow-growing malignancies in our cohort, which are treated in our clinical practice and thus were grouped with other cancers for the present analysis. However, as previously noted, whether treating such malignancies results in improved patient outcomes is unclear. Nevertheless, no demonstrable benefit was observed as a result of using the potentially time-consuming volumetric measurements of the NELSON algorithm. Indeed, measurement of nodule volumes requires specialized software and additional time for nodule segmentation, which may present a barrier to implementing lung cancer screening in some radiology practices. Linear measurements are supported by Lung-RADS and may be obtained with far less effort, and thus their use is advised for routine clinical care.
The present study has several limitations. First, it is a retrospective analysis with a small sample size. In particular, there was a small number of cancers. Second, rather than including consecutive nodules, a sample enriched with nodules assigned to higher categories was generated. However, the bulk of the excluded CT scans were category 1 or 2, and these would have been unlikely to increase the number of cancers in the analysis. In the analysis, percentages were weighted to reflect the distribution of categories across all nodules for which a follow-up lung cancer screening examination was available. Third, because pathologic proof was not obtained for the benign nodules, some such nodules may represent undiagnosed cancers. However, as these patients were in a lung cancer screening program, they would be expected to have received regular follow-up, and any clinically significant lung cancer would likely have been diagnosed during the course of the study. Of note, clinical follow-up was performed for more than 2 years after the final follow-up CT examination in the study sample. Fourth, the exact reasons why particular Lung-RADS categories were assigned at the time of clinical interpretation are unknown. Finally, our analysis was performed within a single health care network that includes two academic sites and one community site. Given the variability and radiologist discretion in aspects of Lung-RADS application, findings may differ at other institutions.
In conclusion, on follow-up lung cancer screening examinations, new nodules had an intermediate cancer risk between stable existing nodules and growing existing nodules. Clinical Lung-RADS showed excellent diagnostic performance for cancer detection on follow-up CT, aside from low specificity in existing nodules. A smaller optimal threshold diameter was identified for new nodules than for existing nodules. Volumetric assessment showed poorer performance than did the clinical Lung-RADS categories. Strict application of Lung-RADS criteria resulted in downgrading of numerous malignant nodules (typically on the basis of stability or pure ground-glass appearance), supporting the use of category 4X for nodules with suspicious features, even if stable. The findings provide insight into radiologists' management of nodules detected on follow-up lung cancer screening examinations.

References

1.
American College of Radiology. Lung-RADS version 1.1. www.acr.org/-/media/ACR/Files/RADS/Lung-RADS/LungRADSAssessmentCategoriesv1-1.pdf?la=en. Published 2019. Accessed June 21, 2019
2.
Xu DM, Gietema H, de Koning H, et al. Nodule management protocol of the NELSON randomised lung cancer screening trial. Lung Cancer 2006; 54:177–184
3.
Walter JE, Heuvelmans MA, Oudkerk M. Small pulmonary nodules in baseline and incidence screening rounds of low-dose CT lung cancer screening. Transl Lung Cancer Res 2017; 6:42–51
4.
MacMahon H, Austin JHM, Gamsu G, et al.; Fleischner Society. Guidelines for management of small pulmonary nodules detected on CT scans: a statement from the Fleischner Society. Radiology 2005; 237:395–400
5.
White CS, Dharaiya E, Dalal S, Chen R, Haramati LB. Vancouver risk calculator compared with ACR Lung-RADS in predicting malignancy: analysis of the National Lung Screening Trial. Radiology 2019; 291:205–211
6.
Hammer MM, Palazzo LL, Kong CY, Hunsaker AR. Cancer risk in subsolid nodules in the National Lung Screening Trial. Radiology 2019; 293:441–448
7.
Horeweg N, van Rosmalen J, Heuvelmans MA, et al. Lung cancer probability in patients with CT-detected pulmonary nodules: a prespecified analysis of data from the NELSON trial of low-dose CT screening. Lancet Oncol 2014; 15:1332–1341
8.
Harris PA, Taylor R, Thielke R, Payne J, Gonzalez N, Conde JG. Research electronic data capture (REDCap): a metadata-driven methodology and work-flow process for providing translational research informatics support. J Biomed Inform 2009; 42:377–381
9.
Hammer MM, Eckel AL, Palazzo LL, Kong CY. Cost-effectiveness of treatment thresholds for subsolid pulmonary nodules in CT lung cancer screening. Radiology 2021; 300:586–593
10.
de Koning HJ, van der Aalst CM, de Jong PA, et al. Reduced lung-cancer mortality with volume CT screening in a randomized trial. N Engl J Med 2020; 382:503–513

Information & Authors

Information

Published In

American Journal of Roentgenology
Pages: 634 - 641
PubMed: 34755524

History

Submitted: October 1, 2021
Revision requested: October 18, 2021
Revision received: October 22, 2021
Accepted: October 31, 2021
Version of record online: November 10, 2021

Keywords

  1. lung cancer screening
  2. Lung-RADS
  3. NELSON
  4. nodule risk

Authors

Affiliations

Mark M. Hammer, MD
Department of Radiology, Brigham and Women's Hospital, Harvard Medical School, 75 Francis St Boston, MA 02115.
Suzanne C. Byrne, MD
Department of Radiology, Brigham and Women's Hospital, Harvard Medical School, 75 Francis St Boston, MA 02115.

Notes

Address correspondence to M. M. Hammer ([email protected]).
The authors declare that they have no disclosures relevant to the subject matter of this article.

Metrics & Citations

Metrics

Citations

Export Citations

To download the citation to this article, select your reference manager software.

Articles citing this article

Media

Figures

Other

Tables

Share

Share

Copy the content Link

Share on social media