NEW YORK (Reuters Health) - Use of the recently proposed Lung-RADS approach to classifying low-dose computed tomography (CT) findings significantly reduces the false-positive result rate, according to a retrospective study.
"With Lung-RADS, the American College of Radiology hopes to standardize the classification and reporting of low-dose CT screening results, in a manner analogous to that with mammography screening and BIRADS, as well as to lower the false-positive rate," Dr. Paul F. Pinsky, from National Cancer Institute, Bethesda, Maryland, told Reuters Health by email. "Using Lung-RADS, the false-positive rate would be expected to be substantially lower than that seen in the National Lung Screening Trial (NLST)."
NLST defined a positive screening result as a nodule of 4 mm or greater in the longest diameter that had no specific benign calcification patterns. Recent reanalysis of NLST data found that increasing the threshold to 6 or 8 mm would have resulted in substantial decreases in the false-positive result rate with only small decreases in sensitivity.
Lung-RADS would increase the size threshold for a positive baseline screening result from 4 mm greatest transverse diameter to a 6 mm transverse bidimensional average (and to 20 mm for nonsolid nodules) and require growth for preexisting nodules for a positive result.
Dr. Pinsky and colleagues used participant- and nodule-level data from the NLST to evaluate the effect of Lung-RADS on the performance characteristics of low-dose CT screening and compared the characteristics of cancer cases detected and missed by Lung-RADS.
For baseline screening, Lung-RADS showed 84.9% sensitivity, significantly lower than the 93.5% for NLST criteria. The false-positive result rate, however, was only half that of NLST (12.8% for Lung-RADS versus 26.6% for NLST), according to the February 10 Annals of Internal Medicine online report.
Results were similar for subsequent annual screenings: sensitivity decreased from 93.8% with NLST to 78.6% with Lung-RADS, while the false-positive result rate declined markedly from 21.8% with NLST to only 5.3% with Lung-RADS.
Lung-RADS would have missed 13% of all true-positive cancer cases: 9.2% on baseline screenings and 16.2% on screenings after baseline. The Lung-RADS stage distribution (stage 1 through stage 4) of missed cancer cases was similar to that of cancer cases that were not missed.
Still, lung cancer-specific survival rates did not differ significantly at five years between the missed cancer cases (71.7%) and the cancer cases not missed (64.2%).
Using Lung-RADS would have reduced the invasive diagnostic procedures by 23% at baseline (through its reduction of false-positive results) and by a similar percentage at subsequent screenings.
"A general mantra of screening is that the false positive rate should be low," Dr. Pinsky said. "This is especially the case with screening for lung cancer where the diagnostic follow-up can involve risky invasive procedures. Also, the anxiety level would be presumed to be high for a positive lung cancer screening test. Therefore, a system that very substantially reduces false positive rates, with corresponding only modest decreases in sensitivity, is a step in the right direction, because false positive rates as high as those observed in NLST are not really sustainable."
"That being said," he added, "the reduction in sensitivity, though modest, is a potential cause of concern, and it will be important going forward to monitor not only the sensitivity rate in large populations undergoing low-dose CT screening, but the possible effects of lowered sensitivity on the mortality benefit of low-dose CT screening."
Dr. Nicole Tripician Tanner, from Medical University of South Carolina's Division of Pulmonary, Critical Care, and Sleep Medicine, told Reuters Health, "I think the more important thing is that a lung cancer screening program utilizes a structured reporting system be it a system based on the NLST criteria or Lung-RADS. This is one of the nine components cited for a high-quality lung cancer screening program and is highlighted in the Center for Medicare and Medicaid Services (CMS) recommendation for lung cancer screening."
"It is very also important that people choosing to be screened for lung cancer understand that they must be committed to a process and not just a single scan done once," Dr. Tanner said. "This is especially true should a higher size threshold be used to declare a screening CT positive; the reduction in sensitivity caused by this approach resulted in missed cancer cases with Lung-RADS making adherence to annual follow-up screening that much more important."
Dr. Christina Bellinger, director of the Lung Screening Program at Wake Forest Baptist Medical Center in Winston Salem, North Carolina, told Reuters Health by email, "Any lung cancer screening should be part of an established program with ongoing data collection so that further improvements in detection rates and false-positive reductions can occur. All programs, in accordance with American Thoracic Society policy recommendations and CMS, should include detailed patient education, primary care physician education, and specialty care involvement (Pulmonology, Radiology, Thoracic Surgery, Oncology, and Radiation Oncology)."
The NLST was funded by the National Institutes of Health. The authors report no disclosures.
SOURCE: http://bit.ly/1MdxUtc
Ann Intern Med 2015.
Δεν υπάρχουν σχόλια:
Δημοσίευση σχολίου