Are the politics of smoking polluting public policy and the fight for low dose CT lung cancer screening?
By Teri Yates
The outlook for people living with lung cancer has always been exceptionally grim. More than half of all patients do not even survive a year after diagnosis, primarily because nearly all lung cancers are detected after the disease has spread to other parts of the body. In its early stages, lung cancer is silent and symptomless, but like so many other cancers when caught in the early stages, it can be treated and in many circumstances it can be cured. The question is, how do we find lung cancer early?
In 1992, physicians at Cornell University Medical Center began looking for the answer to that question, forming the Early Lung Cancer Action Program (ELCAP) to determine if screening with low-dose CT could detect cancer at an earlier stage than chest radiography. ELCAP eventually became an international effort (I-ELCAP), and its promising findings prompted the National Lung Screening Trial (NLST), a randomized clinical trial that examined whether low-dose CT screening could prevent lung cancer deaths in high-risk patients.
Twenty-two years later, the collective work undertaken by these researchers has proven that annual screening with CT is effective at detecting early stage lung cancer, and more importantly that lung cancer screening can prevent 20% of lung cancer deaths. Armed with the necessary evidence demonstrating the effectiveness of screening, researchers and lung cancer advocacy groups have been working diligently to secure insurance coverage for this new and important preventive health service.
Screening Is Only Assured for Population Under 65
Based mostly on the findings of the NLST, in December 2013 the US Preventive Services Task Force (USPSTF) recommended that current or previous heavy smokers between the ages of 55 and 80 undergo annual screening with CT to detect lung cancer. When announcing its recommendation, the task force noted that the data from NLST had been used to conduct extensive modeling studies to identify potential harms and to determine appropriate inclusion criteria and screening intervals. The task force also cautioned that screening should not be undertaken in individuals with other health concerns that would preclude curative treatment in the event of a lung cancer diagnosis.
Under the requirements of the Patient Protection and Affordable Care Act, the USPSTF recommendation has the effect of mandating coverage by insurers, at no cost to patients, by January 1, 2015. This type of recommendation does not automatically trigger coverage for Medicare beneficiaries, however, and under CMS guidelines the agency must also demonstrate that the service is “reasonable and necessary” and that it is specifically appropriate for the population of Medicare beneficiaries. To assist it in making those determinations, CMS referred the topic to the Medicare Evidence Development and Coverage Advisory Committee (MEDCAC). On April 30, MEDCAC shocked lung cancer screening advocates by returning a vote of low confidence, citing concerns about harm to beneficiaries, questions about the applicability of the NLST outcomes in an older population, and doubts about the reproducibility of the outcomes outside the controlled circumstances of a clinical trial.
Ella Kazerooni, MD, chair of the American College of Radiology Lung Cancer Screening Committee, gave testimony during the meeting and was very frustrated by the proceeding. “We [the ACR] are very disappointed. The USPSTF spent 2 to 3 years making a decision to recommend lung cancer screening based on a long thoughtful review. In less than one day, the MEDCAC reached a decision that completely undermined that.” When asked why MEDCAC reached a different conclusion, Kazerooni is quick to point out that the panel evaluated far less material and evidence. “The USPSTF looked at more than just the results of the NLST. They also asked for additional information, for example reviewing the unpublished data accumulated over the past two decades by I-ELCAP. When MEDCAC met, they were only permitted to consider NLST, reviewing nothing that wasn’t in print. They disregarded all of Henschke’s data [Principal Investigator of I-ELCAP].”
Screening Experts Dispute Many of Panel’s Concerns
During the meeting, panelists engaged in an extensive re-litigating of the NLST results. Their concerns were numerous, including:
Underrepresentation of patients over 65 in NLST
- The high false-positive rate, and the psychological distress caused by false-positive results
- The potential harm to patients from radiation or other complications
- The variability in radiologist performance within the trial
- The possibility that patients would be screened outside established guidelines
In addressing issues like adherence to guidelines, panelists and presenters were referencing a broader area of worry, which is whether the outcomes achieved in the NLST can be achieved outside the setting of a clinical trial. For example, Doug Campos-Outcalt, MD, MPH, spoke on behalf of the American Academy of Family Physicians, outlining the group’s position that the evidence is insufficient to recommend lung screening. “Our commission felt that the conditions of the National Lung Screening Trial were unlikely to be replicated in community settings. And we felt there would much less benefit and more harms when this was implemented at a community-wide setting.”
Denise Aberle, MD, the National Principal Investigator of the radiology component of the NLST, advocates for coverage by CMS but also believes the panelist’s reservations are reasonable and should be addressed. “I do have some concerns about quality control and the outcomes that will be achieved outside of the controlled setting of a trial, and I think that CMS could address that by using an approach similar to what they used for oncologic PET scanning,” said Aberle. “When coverage was expanded for new PET indications, they required reporting to a national registry as a condition of reimbursement. This led to the formation of the National Oncological PET Registry (NOPR) to further our understanding of PET. CMS could follow a similar approach for lung cancer screening.”
One way that community-based screening programs can address these types of concerns is to make use of the robust set of tools the ACR has developed to ensure quality and standardization. The ACR has published the Lung Imaging Reporting and Data System (Lung-RADS), which contains assessment categories and management recommendations to promote uniformity in interpretation. The organization also offers an accreditation program for lung cancer screening centers, establishing the technical requirements for the performance of low-dose lung CT and minimum personnel qualifications for those performing and interpreting the studies. The ACR also plans to roll out a national registry for CT lung screening next year.
Kazerooni and Aberle both take issue with the characterization that the findings from NLST only reflect what’s possible in an academic medical center. Aberle explained, “All of the patients were screened at a trial site but quite a few, half or more, received management and follow-up care at other facilities. We provided written recommendations for management to the referring physicians but the care was delivered at a local level.”
Kazerooni also notes that there were no radiologists on the MEDCAC panel, which seemed to contribute to the committee’s misunderstanding of the radiation risks. “The chair, Rita Redberg, MD, a cardiologist, opened with a statement about her work informing the public about the dangers of radiation exposure, which set the tone for the rest of the proceedings.” Redberg also closed the meeting by delivering a final comment in which she estimated the number of cancers she believed would be caused by implementation of screening. Kazerooni’s assertion that several panelists overestimated the potential risks of radiation seems well-supported by the testimony from the only medical physicist presenting at the meeting.
Michael McNitt-Gray, PhD, DABR, medical physicist and chair of the CT Subcommittee of the American Association of Physicists in Medicine, addressed the issue of radiation risk directly in his testimony. McNitt-Gray pointed out that the radiation dose from a low-dose screening CT scan is just half of the radiation dose that Americans receive from background natural sources each year. He went on to compare the cumulative dose that patients receive when undergoing screening (1.5 millisieverts per annual test) to the exposures received by healthcare workers. “Radiation workers such as myself, and radiologists and radiation technologists, are allowed up to 50 millisieverts per year over a 40 year working life.” McNitt-Gray also pointed out that the risks associated with radiation exposure decrease with age, meaning that the potential harm is actually lower for Medicare beneficiaries than younger populations.
In addition to radiation risks, panelists also focused on concerns that other medical complications from screening (sustained during follow-up testing or biopsy) would cause unacceptable harm to beneficiaries. Presenters pointed out that virtually all of these complications occurred in patients who were actually diagnosed with lung cancer, and that the rate of major complications in screening subjects without lung cancer was just a 10th of 1%.
More Cost-Effective Than Other Cancer Screenings
Discussion among panelists about the cost to the Medicare program of screening was fairly limited, which is likely because cost-effectiveness evaluation was not specified as a task of the committee. There was a brief presentation from Bruce Pyenson, an actuary with Milliman Inc, about the cost of the screening the Medicare population. He estimated the cost at $1 per member per month if 50% of the eligible beneficiaries received the service, which amounts to a total cost of $600 million annually or about one-tenth of 1% of Medicare spending.
Pyenson also noted that the dollars per year of life saved is well within the normally accepted range for cancer screenings ($50,000 or less). Lung screening with CT costs $20,000 to $25,000 per year of life saved; in comparison, biennial mammograms for women over 65 cost nearly $37,000 per year of life saved. Pyenson asserted, “This is one of the best valued population interventions I’ve seen, and I think that CMS actuaries with their data would come to the same conclusion.”
The Vote Was About More Than CT Lung Screening
CMS can refer a topic to MEDCAC when there is a lack of consensus among experts about the clinical value and medical necessity of a proposed new service. For lung cancer screening, this is clearly not the case; provision of the service has been recommended by the USPSTF, the American College of Chest Physicians, the American Society of Clinical Oncology, the American Thoracic Society, the American Association for Thoracic Surgery, the American College of Radiology, the Lung Cancer Alliance, the American Cancer Society, and the National Comprehensive Cancer Network.
Throughout the MEDCAC proceeding, there were references to the generalized need for greater caution when approving new screening services, suggesting that regrets over prostate and breast cancer screenings are impacting attitudes about low-dose lung CT. One panelist pointed out that even though the evidence clearly suggests lung cancer screening outperforms those tests, the bar has been raised since their approval. Harry Burke, MD, a professor of bioinformatics and a clinician from Walter Reed National Medical Center, voiced the fear that approval now would be difficult to reverse later. “It’s very important that we don’t get it wrong now, because it will be very hard to get it right later,” said Burke. “Like PSA screening, once it’s in, it’s hard to get it out.”
In its guidance document about when to refer topics to MEDCAC, CMS notes that a referral may be initiated when existing studies contain significant design flaws or studies do not specifically address the special needs of the elderly population. While this is ostensibly the reason for the MEDCAC evaluation, CMS can also seek advice when use of the technology is the “subject of controversy among the general public.” Lung cancer advocacy groups have long argued that the stigma surrounding lung cancer has a significant negative impact on public policy surrounding its diagnosis and treatment. One panelist directly made the point that lung cancer is primarily a self-inflicted disease, arguing that money spent screening for lung cancer would be better directed to encourage people to quit smoking.
“We need to think about if we do cover this, basically you think of it as CMS writing a check for a strategy to reduce deaths from lung cancer that we know are largely caused by tobacco, and year after year the CDC reports significant shortfalls in funding the states for tobacco control efforts. Whether it wouldn’t make sense to allocate our resources directly at tobacco control interventions where we would see absolute risk reduction that would eclipse what we’re seeing with early detection of lung cancer through CT imaging,” said Steven H. Woolf, MD, MPH, from Virginia Commonwealth University. “That’s not to suggest that the important findings reported by the speakers today about how CT screening might encourage people to quit smoking shouldn’t be recognized and applauded, but I wonder if our dollars could go further in actually saving lives from lung cancer by dealing directly with tobacco abuse.”
Aberle has faced this type of opposition before. “I believe that a big part of the issue is reluctance to screen because of the stigma associated with smoking. We saw this when trying to secure funding for NLST—some people were opposed to allocating funds to assist this population. We have to get past this condemnation of people with smoking-related diseases. We do not deny evidence-based treatment for diabetes, heart disease, AIDS, hepatic cirrhosis, or even traumatic injury; yet, many of these conditions have a basis in lifestyle.”
How Blame for Smoking Is Shaping Public Policy
If the stigma of lung cancer played a role in the MEDCAC’s vote, it wouldn’t be the first time that public policy seems to have been influenced by attitudes about smokers. Despite its greater prevalence, lung cancer research is more poorly funded than other common forms of cancer. Withholding research funding—or lung cancer screening—impacts non-smokers and never-smokers as well, who represent 15% of the lung cancer cases diagnosed. Click here for chart: 2012 Federal Research Funding Dollars Per Life Lost.
The American Lung Association has documented and reported the numerous negative effects that the stigma associated with lung cancer has on patients, physicians, and those who fund and research treatment. The great irony of denying lung cancer screening to the Medicare population: Most of the smokers and former smokers within this group took up the addictive habit before the health risks of smoking were widely known.
The first major public warning about smoking risk came 50 years ago with the publication of Smoking and Health: Report of the Advisory Committee to the Surgeon General of the United States. More than 80% of smokers start before the age of 18, which means people who took up the habit before the report are currently older than 67 years of age. And before 1966, there were no health warnings printed on cigarette packaging, with the first specific lung cancer warning not appearing until 1985. This means that the vast majority of beneficiaries who smoke (or formerly smoked) did not receive warnings about lung cancer before starting.
While advocacy groups and most healthcare professionals would reject the idea that any person deserves lung cancer, when it comes to screening the current Medicare population may in fact be the least culpable group of individuals at risk. Click here for table: Lung Cancer Screening Cohort.
MEDCAC Is Not the Final Word on Screening
Despite the vote of low confidence from MEDCAC, the committee’s opinions are just one source that CMS will consider as they make a final determination whether to cover screening. Kazerooni indicates that “we are doing a lot on many fronts at the ACR to help CMS move forward productively and do the right thing by Medicare age beneficiaries so that when they switch from 64 to 65 years, they don’t lose this important health benefit.”
When asked her opinion on the likelihood that CMS will cover screening, Kazerooni is optimistic. “Medicare hasn’t sat in the face of a USPSTF recommendation like this in the setting of the Affordable Care Act to make a decision before, so there is pressure simply from that being there. There is also a lot of pressure in Washington,” she points out. “There were letters from the House and the Senate with a high percentage of sign-on from both parties that went to CMS in support of lung cancer screening CT. I think CMS very much knows that Congress is watching and there is a lot of pressure to expedite a decision.”
Statutorily, the soonest that decision can result in screening for seniors is July 2015. The proposed decision from CMS is due in November 2014, which will allow time for public comment before a final national coverage determination is issued on February 8, 2015. In the meantime, groups like the ACR will continue to engage with CMS and legislators on the issue.
“I think the tone of our conversations with CMS has been very productive recently. The conversation has changed from one of looking at the evidence behind who should get covered to how to roll out coverage responsibly. I think that is a very positive direction,” said Kazerooni.
Teri Yates is Principal Consultant of Accountable Radiology Advisors (ARA), a consulting practice dedicated to helping radiologists and hospitals deliver services of higher value. ARA assists clients with quality program development, virtual quality directorship, and risk management, and also provides value-based contracting, strategic planning, and interim leadership services.
American Lung Association. Addressing the Stigma of Lung Cancer. http://www.lung.org/assets/documents/lung-disease/lung-cancer/addressing-the-stigma-of-lung-cancer.pdf
Centers for Medicare and Medicaid Services Medicare Evidence Development & Coverage Advisory Committee. (2014, April 30). Baltimore.
LUNGevity Foundation. The Truth about Lung Cancer Research Funding: The Government Is Not Focused on Lung Cancer. http://events.lungevity.org/site/PageServer?pagename=v2_Research_TheTruth
The National Lung Screening Trial Research Team. Reduced Lung-Cancer Mortality with Low-Dose Computed Tomographic Screening. N Engl J Med. 2011;365:395-409.
The American College of Radiology. (2014). ACR Webinar: Lung Cancer Screening.
U.S. Preventive Services Task Force. (n.d.) Screening for Lung Cancer: U.S. Preventive Services Task Force Recommendation Statement. AHRQ Publication No. 13-05196-EF-3 . http://www.uspreventiveservicestaskforce.org/uspstf13/lungcan/lungcanfinalrs.htm
United States Centers for Disease Control and Prevention. (2008). Preventing Chronic Diseases: Investing Wisely in Health – Screening to Prevent Cancer Deaths.