A recent report from the U.S. Preventive Services Task Force (USPSTF) has undermined the evidentiary basis for the primary treatment of stage I non–small cell lung cancer (NSCLC) while rating the evidence for the benefit of screening as “high.”1 According to David F. Yankelevitz, MD, these recommendations are contradictory.
During the Quantitative Imaging Workshop XVII, Dr. Yankelevitz analyzed the evidence for surgical resection and stereotactic body radiation in stage I NSCLC and argued that the benefit from screening ultimately comes from early treatment.2 “Surgical treatment of stage I lung cancer is enormously effective, and its evidentiary basis should be considered high,” said Dr. Yankelevitz, Professor of Radiology and Director of the Lung Biopsy Service at Icahn School of Medicine at Mount Sinai, New York. “We all need to understand this treatment is fantastically effective.”
David F. Yankelevitz, MD
As Dr. Yankelevitz reported, draft guidelines for lung cancer screening, prepared by the Agency for Healthcare Research and Quality, found the strength of evidence for the effectiveness of surgical resection and stereotactic body radiation for the treatment of stage I NSCLC to be “moderate” and “low” for benefit, respectively. Per the review, the quality of evidence was downgraded primarily because the evidence came from uncontrolled cohort studies and for imprecision. No randomized controlled trials comparing surgical resection or stereotactic body radiation with no treatment for stage I NSCLC were identified, authors of the review noted.
“This is a breathtaking statement,” said Dr. Yankelevitz. “Lung cancer is the number one cancer killer, and we’re saying there’s moderate to low benefit for the primary forms of treatment. What does that do to trust in what we’re doing?”
Understanding Study Randomization
As Dr. Yankelevitz explained, absence of evidence regarding the usefulness of an experimental intervention can ethically justify randomization, which is referred to as equipoise. However, this principle also mandates that this approach can be applied only when there is a substantial degree of uncertainty as to whether the treatment would benefit study participants. According to Dr. Yankelevitz, the evidence from registries alone is remarkably strong that untreated stage I disease is fatal.
Furthermore, the current staging system is highly dependent on size. If left untreated, a 1-cm tumor with a doubling time of 60 days would be 4 cm at 1 year, and at 2 years (16 cm), death would have already occurred, said Dr. Yankelevitz.
The rationale for performing low-dose computed tomography (CT) trials comes from the National Lung Screening Trial (NLST) itself, he added. The following statements are from the protocol: “A major impetus to migrate from chest-x-ray screening to CT screening for lung cancer is the promise of detecting smaller lung cancers. Yet we do not currently know that outcomes are necessarily better when the cancer is 2 mm as opposed to 20 mm. As purveyors of public policy, we are obliged to avoid the premature endorsement of a screening process before its benefits and liabilities have been reconciled.”3
To answer this question in the simplest, most direct way possible, investigators could randomly assign patients with lung cancer to immediate vs delayed treatment, with an endpoint of overall survival. Combined with CT and chest-x-ray screening to determine the size and stage distribution of early cancers, such a trial would quickly provide more information than either the NLST or NELSON design (lung cancer detection study), according to Dr. Yankelevitz, but it would also be unethical.
“Surgical treatment of stage I lung cancer is enormously effective, and its evidentiary basis should be considered high.”— David F. Yankelevitz, MD
Tweet this quote
To get around this dilemma, Dr. Yankelevitz proposes that investigators randomly assign patients to CT screening vs chest x-ray screening. “Instead of randomizing the treatment side, we should randomize upstream—to the diagnostic test,” he said.
With such a design, patients would be randomly assigned to early diagnosis/early intervention in one arm and symptom diagnosis/late intervention in the other. According to Dr. Yankelevitz, this traditional approach avoids the ethical dilemma of delaying treatment to a person diagnosed with lung cancer.
“Even though we know it is occurring, we would not know which person has the early cancer that is not being treated,” he explained. “Ultimately, both designs would provide the critical endpoint regarding reduction in deaths, in essence randomizing to early and delayed treatment. You just get there a little differently.”
According to Dr. Yankelevitz, despite the assertion of investigators in the draft guidelines for lung cancer screening, there have been randomized trials of surgery for early vs late disease.4,5 “We’ve done this in two remarkably successful trials,” he said. “The NLST and NELSON trials both showed that treatment of early disease is better than treatment of late disease.”
In fact, said Dr. Yankelevitz, based on the results of these trials, the USPSTF now rates the evidence for the benefit of screening as “high.” At the same time, it rates the benefit for surgical treatment and radiation therapy as “moderate” and “low,” respectively.
“The benefit from screening is ultimately from the early treatment that only comes about as a result of early diagnosis,” said Dr. Yankelevitz. “Therefore, it should be acknowledged that, if the evidence for benefit of screening is high, then the evidence for benefit of surgical treatment of stage 1 lung cancer must also be ‘high.’”
According to Dr. Yankelevitz, this is of critical importance. Although the evidence continues to accumulate regarding the benefits of screening, its uptake still remains low. “Reports that undermine the evidence base for the primary form of treatment only raise additional concerns and cast out the entire screening process,” he concluded.
DISCLOSURE: Dr. Yankelevitz holds stock or other ownership interests in Accumetra; has served as a consultant or advisor to GRAIL; and holds patents or other intellectual property “relating to the evaluation of diseases of the chest including measurement of nodules.”
1. Jonas, DE, Reuland, DS, Reddy SM, et al: Screening for Lung Cancer with Low-Dose Computed Tomography: An Evidence Review for the U.S. Preventive Services Task Force. Available at https://www.uspreventiveservicestaskforce.org/uspstf/. Accessed on November 17, 2020.
2. Yankelevitz DF: Evaluating the effectiveness of surgical resection of stage I lung cancer. Quantitative Imaging Workshop XVII. Session 6. Presented October 30, 2020.
3. Aberle DR, Black WC, Goldin JG, et al: Contemporary screening for the detection of lung cancer protocol [NLST], 10 May 2002. American College of Radiology Imaging Network (ACRIN sharp6654).
4. 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 382:503-513, 2020.
5. Duffy SW, Field JK: Mortality reduction with low-dose CT screening for lung cancer. N Engl J Med 382:572-573, 2020.
Matthijs Oudkerk, MD, Professor of Radiology at the University Medical Center Groningen, Netherlands, and Principal Investigator for Radiology of the NELSON Lung Cancer Detection Study, stressed that detecting individuals who have early-stage lung cancer with more virulent biologic behavior is...