Delays in Follow-up after a Positive Lung Cancer Screening Exam: Is the Benefit of Screening Compromised?

Following the results of several randomized trials, we can agree that the benefits of lung cancer screening (LCS) with low-dose computed tomography (LDCT) in high-risk individuals are indisputable (1, 2). A recent meta-analysis of nine trials including 88,497 patients showed that LCS is associated with a significant reduction in lung cancer mortality, with an overall risk ratio (RR) of 0.87 (95% confidence interval [CI], 0.78–0.98) compared with no screening or chest X-ray and an RR of 0.80 (95% CI, 0.69–0.92) compared with no screening. The benefit in lung cancer mortality reduction associated with screening is related to a significant increase in the diagnosis of early-stage lung cancer overall (RR, 2.84; 95% CI, 1.76–4.58), compared with no screening (RR, 3.33; 95% CI, 2.27–4.89), and compared with chest X-ray (RR, 1.52; 95% CI, 1.04–2.23) and a significant simultaneous decrease in the diagnosis of advanced-stage lung cancer overall (RR, 0.75; 95% CI, 0.68–0.83) and compared with no screening (RR, 0.67; 95% CI, 0.56–0.80) (3). Multiple steps across the LCS pipeline are required to realize the benefit of early lung cancer detection and reduced lung cancer mortality. Although each step is critical, adherence to annual follow-up after a negative screening examination and to recommended follow-up and diagnostic evaluation after a positive screening examination is essential.

The American College of Radiology (ACR) Lung CT Screening Reporting and Data System (Lung-RADS) is a quality-assurance tool developed to standardize the reporting of LDCT, provide management recommendations after a positive examination, and facilitate monitoring of screening outcomes (4). To manage positive examinations, the ACR provides specifically recommended timelines for follow-up LDCT and additional imaging studies for Lung-RADS category 3 (6 mo) and category 4A (3 mo) findings and recommends immediate follow-up with diagnostic imaging studies or biopsy for Lung-RADS category 4B or 4X findings (4). A current limitation of Lung-RADS is the lack of a clear recommended timeline for diagnostic procedures and tissue sampling for “very suspicious” Lung-RADS 4B or 4X findings (5).

Implementing LCS has been fraught with leaks along the pipeline, including low follow-up rates after a positive LCS examination. A meta-analysis reported a pooled adherence rate of 74% for Lung-RADS category 3 and 4 findings but highlighted the variability of follow-up rates as a result of heterogeneity in how recommended follow-up time after a positive LCS examination was defined across studies (6). In studies reporting adherence rates to the ACR-specific recommended follow-up time after a positive examination (Lung-RADS category 3, 4A, 4B, or 4X), overall rates are even lower, between 42.6% and 63.1% (5, 7). Although adherence rates improve with higher Lung-RADS assessment category and when the follow-up timeline is extended beyond the ACR Lung-RADS guidance (5, 7), delays beyond the recommended follow-up care raise concerns about the potential compromise of the benefits of LCS (5).

In the retrospective study accompanying this editorial in this issue of AnnalsATS, Ahmed and colleagues (pp. 1175–1181) evaluated delays in follow-up after a positive LCS examination (defined as Lung-RADS category 3, 4A, 4B, or 4X) among 369 patients enrolled in a multisite LCS program in the Seattle area (8). The authors defined the recommended follow-up interval as follow-up 30 days beyond the ACR-recommended timeline (for Lung-RADS category 3, 6 mo + 30 d; for category 4A, 3 mo + 30 d; and for categories 4B and 4X, 30 d) and defined delayed follow-up as the time to follow-up exceeding the study’s defined time intervals, death, or censoring. Overall delays in follow-up were found in 47% of positive screening examinations, with a median delay of 104 days. Like prior studies (5, 7), longer delays were found among those with Lung-RADS category 3 findings (59%; 210 d) compared with category 4A and 4B or 4X findings (35% [64 d] and 40% [34 d], respectively; P < 0.001). Of the 54 patients diagnosed with non–small-cell lung cancer (NSCLC) in whom clinical stage at follow-up could be identified, 16 (30%) had a delay in follow-up and 14 (26%) had upstaging (defined as any increase in T [tumor], N [node], or M [metastasis] stage between index LDCT and earliest follow-up). Delays in follow-up were associated with an increased likelihood of clinical upstaging from the LCS examination to the follow-up examination (P < 0.001); the extent of delay was also significantly greater in those with any clinical upstaging (n = 11 of 16) than in those without any upstaging (n = 5 of 16) (median, 131 d vs. 15 d; P = 0.047). Of the 11 patients with delayed follow-up and upstaging, 2 had Lung-RADS category 3 findings, 3 had category 4A findings, and 6 had category 4B or 4X findings at the LCS examination. Upstaging was primarily driven by an increase in the size of the primary tumor (i.e., T stage) (12 of 14 with any upstaging) and occurred mainly in patients with Lung-RADS categories 3 and 4A findings at the LCS examination. In contrast, clinically significant upstaging driven by increased nodal involvement (i.e., N stage) was seen only in patients with Lung-RADS category 4B or 4X findings at the LCS examination (8).

Although the findings of delays in follow-up after positive LDCT findings are not new, the impact of delays in follow-up care after a positive LCS examination on lung cancer upstaging reported by Ahmed and colleagues is novel and clinically meaningful. Staging in NSCLC is crucial because it determines treatment and prognosis. As noted, the benefit of LCS stems from a significant increase in the detection of stage I NSCLC, which is associated with the best prognosis (9). The finding of significant upstaging driven by N stage in the study by Ahmed and colleagues is most concerning because it results in a potentially significant change in the stage and subsequent treatment that is often associated with poorer patient outcomes. The finding of upstaging by T status is also of importance. The eighth edition of the TNM classification of lung cancer modified the T classification based on 1-cm increments in size. Survival is inversely proportional to every 1-cm increase in tumor size (9).

Delay before the first follow-up after a positive examination does not factor in additional delays or barriers to surgery or other treatment for lung cancer, which may further compromise the benefits of LCS. Outside of LCS, several studies have explored the relationship between the timing of surgery following the diagnosis of stage I NSCLC and overall survival, with mixed results. For example, some studies suggest that adverse patient outcomes are associated with delayed surgery for early-stage NSCLC (10–12), yet others have shown no association (13–15). The findings across these studies are difficult to compare and synthesize as a result of considerable heterogeneity in their study designs, populations, tumor histologies, and methods, including the definition of time to surgery.

Although, in the future, we will likely rely on complementary technologies, including genomic biomarkers, exhaled-breath analysis of volatile organic compounds, and artificial intelligence, to improve the diagnostic accuracy of pulmonary nodules detected on LCS examinations and shorten the time to diagnose lung cancer, we are not there yet. Therefore, we must aim to develop and implement strategies to improve adherence to the ACR-recommended time to follow-up after a positive examination and standardize recommendations for the timing of diagnostic testing of very suspicious Lung-RADS category 4B or 4X findings to avoid compromising the benefits of LCS.