Detecting Lung Cancer: Is This A Better Way?
—A lung cancer screening test examining small noncoding RNAs has demonstrated the ability to identify lung cancer using whole blood collected in the clinical setting, making it a potential alternative to low-dose computed tomography.
Lung cancer screening with low-dose computed tomography (LDCT) has been used as a screening test to detect cancer at an earlier stage and has been associated with reduced mortality.1,2 As a result, annual lung cancer screening with LDCT is now recommended in the U.S. for people ages 50 to 80 years old with at least a 20 pack-year smoking history who currently smoke or who quit within the past 15 years.3 Lung cancer screening has not been implemented as widely in Europe, however.4
Yet despite recommendations for lung cancer screening with LDCT, uptake has been limited. In the U.S., only 4.5% of those eligible were screened in 2022.5
With sluggish use of screening in mind, researchers in the U.S. and Germany hypothesized that “a blood test, if integrated into routine annual checkups within primary care, could significantly lower the barrier to participation in lung cancer screening.”4
Thus, the research team aimed to identify a biomarker from peripheral blood that could serve as a lung cancer screening test. They utilized an RNA-based platform that overcomes some disadvantages of liquid biopsy approaches that rely on the detection of circulating tumor DNA or circulating tumor cells. Liquid biopsy approaches may have value in late-stage disease, but its diagnostic sensitivity is low—roughly 20% to 50%—for detecting early-stage lung cancer, “primarily due to low shedding rates in early-stage tumors,” the authors stated.4
Developing a blood-based screening test
In an interview with Ƶ, the study’s corresponding author, Bruno R. Steinkraus, PhD, with Hummingbird Diagnostics GmbH, in Heidelberg, Germany, explained that “historically, small RNAs have suffered from pre-analytical instability, largely caused by open pipetting and sample handling in the clinic.” To address this limitation, Dr. Steinkraus says the research team “implemented a ‘pipetting-free’ approach, in which sample handling is moved to the controlled lab environment, where highly abundant red blood cell small RNAs are effectively removed from the library preparation after cell lysis.”
This prospective study, which was published in the Journal of Thoracic Oncology, was performed in collaboration with multiple U.S. and German medical centers.4 At the U.S. medical centers, participants were identified from among those who were scheduled to undergo LDCT. In Germany, where LDCT lung cancer screening has not yet been implemented, participants were recruited from among people suspected of having lung cancer or among those seeking pulmonary care at a tertiary lung hospital. Lung cancer diagnoses were determined based on chest CT findings in Germany, or by biopsy, surgery, or more than 12 months of follow-up in the U.S.
A total of 1384 participants were included. After using collection tubes to lyse the samples and stabilize cellular and extracellular RNA, Small RNA (sRNA) sequencing and blocking of highly abundant erythroid-derived microRNAs (miRNAs) were performed. Next, polymerase chain reaction (PCR) artifacts were removed and sRNA types were categorized, as certain sequences were removed. Logistic regression models were then trained and validated to form a model that could identify sRNAs associated with lung cancer.
Cell sorting was performed using samples from 21 participants with noncancer lung disease and 19 participants with lung cancer. Flow cytometry measured the yield, quality, and purity of isolated cell populations. For each blood sample, cell counts were determined for erythrocytes, thrombocytes, leukocytes, neutrophils, eosinophils, monocytes, T cells, CD4+ T cells, CD8+ T cells, B cells, and NK cells. Cellular origins of RNA profiles were then determined.
What the investigators accomplished
The researchers were successful in blocking abundant erythroid miRNAs allowing next-generation sequencing (NGS) could be used to improve detection of sRNAs from plasma and immune cells.
Test performance varied by lung cancer stage and ranged from areas under the curve (AUCs) of 0.73 for stage I disease to 0.90 for stage IV disease. Detection rates were superior for squamous cell carcinomas (AUC 0.88) compared with adenocarcinomas (AUC 0.81). The AUC for detection of small cell lung cancer was high, at 0.93. Overall, current smokers (AUC 0.86) were more likely to have detectable disease than former smokers (AUC 0.82).
The research team was also successful in developing and validating an 18-sRNA feature consensus signature that they refer to as “miLung,” which is comprised of both tumor-derived and cellular features that differentiate persons with lung cancer from those without lung cancer.
A novel 28S rRNA fragment, 28S-rRNA-bin-162, was identified as the most strongly predictive of lung cancer. Its expression increased with pathologic stage. Furthermore, repeated blood analyses following tumor resection demonstrated stage-dependent decreases in the level of 28S-rRNA-bin-162 (-3% for stage I, -33% for stage II, and -51% for stage III).
A novel approach to lung cancer screening
“The principal finding of our research is that blood-based small noncoding RNAs are useful and accessible biomarkers that can enable highly accurate lung cancer detection,” explained Dr. Steinkraus.
In addition to allowing “ultra-deep investigation of small RNA candidates that had previously been masked from NGS detection,” Dr. Steinkraus says the team showed that “small RNAs used in our test can be readily detected from dried blood spots collected from capillary blood, suggesting a potential future home-sampling application.”
In contrast to LDCT screening, which requires multiple steps and contacts with the healthcare system, “the blood-based miLung test could be performed during the annual physician visit, at the time other blood is being drawn, without the need to schedule additional appointments,” Dr. Steinkraus says, adding that “this will dramatically increase participation in lung cancer screening and thereby have a substantial impact on lung cancer mortality.”
Following the promising results from this study, Dr. Steinkraus and his colleagues plan to conduct further prospective studies in the U.S. and Germany to evaluate the performance of miLung compared with other blood-based lung cancer screening tests.
Published:
References