With Medicare on the cusp of a momentous decision about whether to pay for CT lung cancer screening, the issue is the focus of a pair of editorials published October 13 in JAMA Internal Medicine. Taking opposite sides, the articles one last time outline positions that have become solidified as the U.S. government weighs what could be one of the biggest public health initiatives in decades.
The U.S. Centers for Medicare and Medicaid Services (CMS) is expected to approve or deny screening for Medicare patients at high risk for lung cancer as soon as next month. The stakes couldn't be higher, with proponents urging the government to pay for an exam that already has shown a better cost-benefit ratio than other tests such as breast screening, and which could save tens of thousands of lives.
Meanwhile, detractors claim that CT lung cancer screening hasn't been adequately studied, with only one large clinical trial demonstrating its utility. Paying for the scans at this stage could open a Pandora's box of issues, ranging from higher healthcare costs to a sharp increase in harms from screening, they claim.
High stakes
The editorials are essentially a follow-up to the controversial meeting in April of the Medicare Evidence Development and Coverage Advisory Committee (MEDCAC). The panel was expected to sign off on Medicare coverage of screening for older smokers, but instead it delivered a stunning setback to screening proponents by issuing a vote of low confidence in the exams.
In explaining its decision, the MEDCAC panel cited concerns that evidence backing CT lung screening was insufficient: in particular, that screening's harms could outweigh its benefits in older people, a position that screening advocates say disregards substantial evidence favoring screening. Screening proponents charge that this week's editorials are little more than an effort to revive arguments that have already been refuted by the weight of evidence and by solid preparation for screening going forward.
The new editorials outline many of the positions that were discussed at the meeting. The editorials are introduced by Dr. Robert Steinbrook, a member of the MEDCAC panel from 2010 to 2012. Dr. Douglas Wood, from the University of Washington, argues on the proscreening side, while Dr. Steven Woolf, from Virginia Commonwealth University, and colleagues argue against screening.
In his introduction, Steinbrook noted that the CT screening decision is among the most consequential and closely watched coverage determinations that CMS has made in many years.
"At issue is not only whether low-dose CT will be covered but, if it is covered, the specifics, such as the number and frequency of scans, the beneficiaries who would be eligible, and the procedures to assure that scans are of high quality and that false-positive results are minimized," Steinbrook wrote (JAMA Intern Med, October 13, 2014).
Wood: Screening saves lives
Wood noted that the National Lung Screening Trial (NLST) found a 20% mortality reduction in long-term smokers at high risk of lung cancer, defined as individuals with a minimum 30-pack-year history of smoking. This led the U.S. Preventive Services Task Force (USPSTF) in 2013 to recommend CT lung cancer screening, granting a "B" recommendation.
That set in motion a requirement that private insurers pay for the scans for their beneficiaries younger than age 65 under the Affordable Care Act. However, USPSTF's recommendation doesn't apply automatically to Medicare recipients unless CMS deems it worthy in this population, triggering the current debate between the pro and con forces.
Existing evidence is compelling on the question of whether the experience of 53,000 NLST screening subjects is sufficient to begin widespread screening in seniors in addition to other age groups, Wood believes. Considering that 70% of lung cancers occur in individuals older than 65, it makes no sense to cut off screening at a time when the majority of lung cancers are expected to occur, he wrote.
"It is important to avoid the situation of at-risk patients being screened up to age 64 years through private insurers and then abruptly ceasing screening at exactly the ages when their risk for developing lung cancer is increasing," he wrote.
Lung cancer is the second most common cause of death in the U.S. after heart disease, and it is the most common cause of cancer death, exceeding the next four cancers combined (breast, colon, pancreas, and prostate), he continued.
Also, because lung cancer rarely becomes evident before it has reached an advanced, incurable stage, it must be treated earlier than the other four cancers to save significant numbers of lives, Wood wrote. More than 70% of the next four most common cancers can be treated at a local or regional stage, compared with just 37% of lung cancers.
An analysis of data from NLST found that screening smokers at high risk of lung cancer can reduce mortality by at least 16%. More important to the question at hand, NLST demonstrated that the "mortality reduction was similar for patients younger than 65 years or older than 65 years and for current or former smokers," Wood wrote.
Screening's critics warn of unintended consequences, including risks inherent in imaging follow-up, biopsy, and surgical intervention. However, in NLST, lung cancer screening produced significant benefits after accounting for the harms.
Having to screen 320 patients to save one life from lung cancer might seem like a lot of patients, Wood wrote, but the number needed to screen is much larger for other common screening tests such as breast cancer (2,000 people) or colon cancer (1,200 people).
What's more, the new LUNG-RADS reporting criteria from the American College of Radiology (ACR) decreases the positive screen rate by more than 50%, according to Wood. In addition, ACR has developed comprehensive criteria for screening centers to ensure consistently high scan quality -- an issue raised during the April MEDCAC meeting.
As for ensuring consistent screening quality, patient advocacy group the Lung Cancer Alliance has developed a centers-of-excellence program that helps identify facilities with the required expertise to maximize the benefit of screening. And for its part, the Society of Thoracic Surgeons has published standards for the surgical management of patients with lung cancer, and it tracks patient outcomes in its registry.
"A decision to add lung cancer screening with low-dose CT as a covered benefit for Medicare beneficiaries should be inevitable given the high level of evidence that screening can lead to early diagnosis and cure for thousands of patients each year in the U.S.," Wood concluded. "Professional societies have great expertise and experience in screening and can help CMS responsibly implement a program that is patient-centered and minimizes unintended harms and costs."
Woolf: How strong is the evidence?
In his editorial, Woolf stated that the majority of evidence supporting screening came from a single trial: NLST. It is uncertain whether that experience will extend from the carefully controlled academic trial setting to routine clinical practice.
"The potential harms -- which could affect a large population -- include false-positive results, anxiety, radiation exposure, diagnostic workups, and the resulting complications," he wrote. "It is unclear if routine screening would result in net benefit or net harm. The NLST may not be generalizable to a national screening program for the Medicare age group because 73% of NLST participants were younger than 65 years."
Woolf also cited risks of overdiagnosis -- i.e., of detecting and removing tumors that if left in place would never have harmed the patient. One analysis of NLST data found a rate of 1.4 such diagnoses for every cancer averted.
On the positive side, screening averted 83 deaths from lung cancer, he wrote. On the negative side, screening led to 16 iatrogenic deaths from diagnostic workups, which included 10,246 imaging studies, 322 percutaneous biopsies, 671 bronchoscopies, 713 surgical procedures, and 228 complications (86 classified as major), Woolf asserted.
Then there is the tendency of screening advocacy organizations to expand the list of potential screening subjects to those at lower risk of cancer, Woolf wrote. Already, organizations are proposing to reduce the minimum age at which screening should start to 50 rather than 55 years, to reduce smoking history from 30 to 20 pack years, and to include smokers who quit more than 15 years ago. Including these patients would lead to more false positives but few additional cancers, according to Woolf.
If the criteria were changed to end the limits on the number of annual screening rounds, a 55-year-old smoker who continued to smoke could receive up to 22 additional follow-up scans by age 80, potentially exposing these individuals to risks from radiation, he wrote.
"In our view, the best course of action is for CMS to postpone a coverage decision until better data become available, such as evidence anticipated from ongoing European trials, a [Veterans Affairs] pilot program, and research on harms in everyday practice," Woolf wrote. "The absence of Medicare coverage could create a two-tiered situation, given the mandated coverage of screening by commercial plans. As an ethical matter, however, gaps in the data make it unclear whether postponing a decision would deny benefit or spare harm for Medicare beneficiaries."
Screening proponents disagree
The concerns expressed by the antiscreening side in the editorials focus on the unknown balance of benefits and harms; however, screening in NLST was completed more than a decade ago and did not contain substantial new information about benefits and harms, according to Dr. David Yankelevitz, a professor of radiology at the Icahn School of Medicine at Mount Sinai.
"While the NLST was primarily performed at academic institutions, the associated workups did not have this requirement, and there were no requirements to follow a protocol," Yankelevitz said. "The idea that newer, more efficient, evidence-based protocols could not be followed ... is unreasonable."
Also, it's misleading to say that the benefits and harms of screening in the older population aren't understood simply because there were fewer older people in NLST.
"This represents a very serious misinterpretation of what the NLST actually demonstrated," Yankelevitz said. "Its core result showed that screening with CT finds smaller, more curable cancers than chest x-ray. It does not speak to the magnitude of the benefit. Nor does it speak to a particular age where somehow cancers might behave differently. There is nothing to suggest that cancers found early in 65-year-olds are any less curable than those found in the younger ages in NLST."
Lung cancer researcher and clinical professor of radiology Dr. Claudia Henschke, PhD, also from Icahn School of Medicine, noted that an analysis of participants ages 65 and older in the International Early Lung Cancer Action Plan (I-ELCAP) showed that CT screening produced the same benefits as it did in younger participants ages 55 to 74.
"Potential harms can be minimized in all age groups, including those 65 and older, by following a well-defined evidence-based regimen of screening," Henschke wrote in an email to AuntMinnie.com. "Also, the core result of the NLST extends to continued annual rounds of screening and should not be interpreted as limited to three rounds; this is a serious misunderstanding of the NLST design."
'Half-truths and insinuations'
The debate roils as both sides accuse the other of playing politics. In his editorial urging that screening be approached cautiously, Steinbrook wrote that industry groups are strongly advocating screening, aided by members of Congress, nearly 200 of whom signed a letter in June urging the adoption of low-dose CT screening. Late last month, more than 70 organizations and medical societies similarly urged CMS to adopt screening for seniors at risk.
Meanwhile, screening proponents have pointed out that the editor of JAMA Internal Medicine, Dr. Rita Redberg, also chaired the April MEDCAC meeting. Redberg, a cardiologist and professor of medicine at the University of California, San Francisco, is well-known to the radiology community, having penned a recent editorial on the risks of CT radiation dose, as well as an opinion piece praising the rejection of Medicare coverage for virtual colonoscopy in 2009.
Laurie Ambrose Fenton, president of the Lung Cancer Alliance, said that today's editorials rehash "previously raised -- and rebutted -- half truths and insinuations on lung cancer screening risks."
In an email to AuntMinnie.com, ACR echoed those criticisms, charging that the editorial "ignores available evidence, relies on unsubstantiated or ambiguous claims, and fails to accurately portray the current state of lung cancer screening."
NLST showed that lung cancer screening substantially reduced lung cancer deaths, and nearly one-third of trial participants were age 65 or older, ACR wrote, citing the recent analysis of NLST results in Medicare-aged patients showing that screening works as well in older screening subjects as it does in younger ones.
Given that lung cancer kills 160,000 people each year, "CT lung cancer screening can save more people than any single cancer test developed to date," ACR wrote.
Dr. James Mulshine, professor of internal medicine and dean of the graduate school at Rush Medical College, said the issue of CT lung cancer screening has been debated thoroughly and is now settled.
"USPSTF has recommended [and] professional societies and advocacy groups have overwhelmingly endorsed" lung cancer screening, he wrote in an email to AuntMinnie.com. "The urgent priority now is to focus this consortium on the challenging task of national implementation, so that every high-risk individual in the target population has access to safe, effective, and economical screening service so we can reduce the mortality of the world's most lethal cancer. It's time to move forward and do this right."
