The Great Debate: Investigating the Utility of Low-dose CT Lung Cancer Screening & Reimbursement

A big question mark sits over the topic of low-dose CT lung cancer screening. Since the April meeting of the Medicare Evidence Development & Coverage Advisory Committee (MedCAC), many are left wondering about the controversy over low-dose CT lung cancer screening (LDCT) and the eventual decision that will be made regarding Medicare reimbursement for screening. What does the research really reveal about its effectiveness, and what do proponents and opponents to reimbursement say about the contentious topic?

Back in December 2013, the U.S. Preventive Services Task Force (USPSTF) recommended annual screening with LDCT for asymptomatic individuals between the ages of 55 and 80 who have a 30 pack-year history of smoking and currently smoke or have quit within the last 15 years. The recommendation, labeled as grade “B,” stated screening be discontinued once a patient has ceased smoking for 15 years or develops a health problem that drastically limits his or her life expectancy or willingness to undergo curative lung surgery. With a grade “B” status, private insurers must cover the procedure without a co-pay under the Affordable Care Act. The healthcare reform law does not, however, force Medicare to provide full national coverage for the recommendation. Here enters the conflict. 

The USPSTF’s recommendation was largely based upon the National Lung Screening Trial (NLST), which found that annual LDCT scans could reduce the chances of dying from lung cancer by up to 20 percent in the high-risk patient population. The trial enrolled 53,454 current or former heavy smokers from 33 sites and coordinating centers across the United States. 

Participants in the trial were randomized to three annual screens with either low-dose helical CT or a single-view chest X-ray. Lung cancer incidence per 100,000 person-years was 645 in the LDCT arm and 572 in the chest X-ray arm. Lung cancer mortality was 247 per 100,000 person-years in the LDCT arm and 309 in the chest X-ray arm, demonstrating a 20 percent reduction in lung cancer mortality in the LDCT arm. All-cause mortality was reduced by 6.7 percent for participants who underwent LDCT compared to those who got X-rays. 

On average during the three rounds of screening, 24.2 percent of the LDCT screens were positive and 6.9 percent of the chest X-rays were positive. In both arms, the majority of positive scans led to more screening. Across all rounds, the false-positive rate was 96.4 percent for the LDCT tests and 94.5 percent for the chest X-rays. 

Considering all pieces of the screening puzzle

While the results of the NLST show promise for reducing mortality among this high-risk population, concerns loom over overdiagnosis. A study led by Edward F. Patz, Jr., MD, of Duke University School of Medicine in Durham, N.C., examined data from the NLST to estimate overdiagnosis in the cancers detected by LDCT.

The study’s results indicated approximately 20 percent of all the lung cancers found were attributed to overdiagnosis; these cancers would never have affected the patient’s outcome. Patz attests that overdiagnosis should be one limitation to consider when patients are deciding whether or not to undergo screening.

“In the future, once there are better biomarkers and imaging techniques to predict which individuals with a diagnosis of lung cancer will have more or less aggressive disease, treatment options can be optimized, and a mass screening program will be more efficient and effective,” wrote Patz and colleagues in the study, which was published online December 9, 2013, in JAMA Internal Medicine.  

Patz says, however, that no strides ready to be implemented in clinical practice have been made in this realm yet. “I think this is an area that we need more resources and research in to really understand which cancers have a fairly benign phenotype and will not likely affect the patient,” he says. 

In terms of the study’s implications on the future of LDCT, Patz remarks that overdiagnosis is one potential limitation of any screening test. 

“Screening will almost always find some disease that may not become clinically apparent,” says Patz. “The more you interrogate, the more you’re going to find abnormalities that you’re not going to know what to do with.” 

To craft a public policy regarding screening, Patz suggests considering all pieces of the puzzle involved, including overdiagnosis and false-positives. “It appears that one of the MedCAC concerns was that there are consequences every time we use the technology. Both sides of the equation need to be addressed in order to determine the full impact of screening. LDCT impacts lung cancer mortality, but is expensive and has limitations, and we need to make sure that we are doing the right test for the patients and provide full disclosure of the risks and benefits,” says Patz. 

But what will it all cost? A lot. A recent study utilizing a simulation model estimated that implementing the USPSTF’s lung cancer screening recommendations would likely double the proportion of lung cancers found at an early stage, but at considerable cost, according to the study’s lead author, Joshua Roth, of the Fred Hutchinson Cancer Research Center in Seattle. 

“LDCT screening will cost a substantial amount,” says Roth. “This is because the additional cost of screening millions of Medicare enrollees will greatly outpace any potential cancer care savings from a stage shift. Accordingly, CMS will need to find cost offsets to balance their budget, or will need to increase premiums to fund the implementation of lung cancer screening.”

Using a mathematical model that calculated the cost of the screening over a five-year period and was based on three scenarios, Roth and colleagues estimated doctors would order 11.2 million more CT scans, resulting in nearly 55,000 more lung cancers detected over five years vs. no screening. This estimation was made under the most likely scenario, which assumed half of people who are offered screening would get tested. The program, which would include the cost of screening exams, diagnostic work-up and care of newly detected cancers, is estimated to cost Medicare $9.3 billion over five years. This translates into $3 on a per Medicare enrollee per month basis.

The researchers also estimated the program would more than double the proportion of cancers diagnosed at an early stage, with an increase from 15 to 33 percent. 

“LDCT screening has a potential to precipitate a major shift in lung cancer diagnosis to earlier, more curable stages, even after accounting for sub-optimal ‘real world’ screening uptake,” explains Roth. “This is a critical point because it means many lives can be saved or extended by implementing LDCT lung cancer screening. This also has specific implications for surgical specialists and radiation oncologists, who may find rapidly increasing demand for early-stage lung cancer-related procedures over the coming five years.”

Making sense of the MedCAC meeting

During the MedCAC meeting, representatives from organizations like the American College of Radiology and the Lung Cancer Alliance argued for national Medicare coverage of LDCT for high-risk patients. Although the Centers for Medicare and Medicaid Services (CMS) will not post a draft coverage decision until November, the MedCAC panelists did not have high confidence that current evidence supported the notion that the benefits of LDCT lung cancer screening in the Medicare population would outweigh the harms. 

One of those people lacking in assurance was Doug Campos-Outcalt, MD, MPA, of the University of Arizona College of Medicine in Phoenix. Campos-Outcalt spoke at MedCAC on behalf of the Academy of Family Physicians and listed five major concerns with the USPSTF recommendation. The AAFP ultimately contended that evidence is insufficient for recommending for or against LDCT. The AAFP’s five major concerns stemmed from the recommendation being based largely on one study; the ideal conditions of the NLST that are unlikely to be reproduced in the community setting; extending the number of screening tests beyond the NLST by modeling; the possibility of a current smoker undergoing 25 annual CT scans and the lack of cost benefit analysis. 

When speaking of the MedCAC decision, Campos-Outcalt says, “I think they’re right to be a little cautious. I thought they were asking good questions and were pretty astute in looking at the issue.”

He adds: “Ultimately, it’s about where you want to put your values. If you’re concerned about harms when implementing the screening community wide, then you need to be more conscious.” Campos-Outcalt points to the NLST’s patient population, noting that about two-thirds were under the age of 65 and tended not to have many comorbid conditions. He believes that more harms from biopsies, procedures and studies would be found in older participants with more comorbid conditions.

“There were smaller studies done in Europe and while the designs were not as robust as the NLST, they did not show as much benefit,” says Campos-Outcalt. “When it comes to overall mortality, some didn’t show a difference. It’s interesting because when you do a meta-analysis and include the NLST’s data, you in fact get very little difference in overall mortality.” 

Given the debated reduction in mortality found in the NLST, Campos-Outcalt believes that smoking cessation programs would likely be a more beneficial alternative to screening. “In the best of conditions, the reduction in the mortality rate from screening is 20 percent; 80 percent of the mortality is still there. If we end up with more harms, it won’t take many to negate the benefit of screening,” he says. 

While Campos-Outcalt does not support Medicare reimbursement for LDCT given the current evidence, Ella Kazerooni, MD, of the University of Michigan Health System in Ann Arbor and chair of the American College of Radiology’s Committee on Lung Cancer Screening, sees it differently.   

“I think it was an unfortunate decision made by a group of people who were not allowed to consider all the important information needed to make an informed decision,” she says. “I don’t think they used the correct focus. Whenever we tried to educate the panel about the checks and balances that are already in place today for CT, we weren’t allowed to talk about those things.” 

So will the decision be based on one study? Yes, but with good reason, Kazerooni asserts. “This one study was the NLST, a study that spanned over eight years, included over 50,000 people, was backed by $240 million plus tax payer dollars through the National Cancer Institute, and demonstrated a significant reduction in death from lung cancer. If the message is that we’re not going to make a coverage decision based on that quality of evidence, then we’re sending the wrong message about clinical research. It’s saying that tax payer dollars were basically a waste.”

Kazerooni also rebukes assertions that the trial was done in ideal settings and does not reflect outcomes in community practices. “The assumption was made during the meeting that all follow-up imaging and downstream testing were done at sites for the NLST. This wasn’t true; most participants went back to their local primary care doctors. The NLST outcome was achieved in spite of not mandating certain follow-ups, but by care delivered by physicians in local communities nationwide.” 

Kazerooni believes attention needs to shift toward quality, safety and outcomes in practice moving forward. “Sometimes when you feel like you have a setback, it’s actually a gift,” she says. “In many ways, I see what happened at MedCAC as a gift. We want to show CMS what is being done and what can be done to ensure that the screening is safely and effectively used in clinical practice. We want to give CMS the information that they need to maximize the benefits and minimize the harms of LDCT. We want to help CMS do the right thing.”

A draft coverage decision will be posted on November 10th. Stay tuned