2 Clarke Drive
Suite 100
Cranbury, NJ 08512
© 2024 MJH Life Sciences™ and OncLive - Clinical Oncology News, Cancer Expert Insights. All rights reserved.
Inga T. Lennes, MD, MBA, MPH, discusses the recommended screening protocol for patients at risk for lung cancer, maintaining patients post-screening, and how physicians can best assist this patient population.
Inga T. Lennes, MD, MBA, MPH
Patients who are between the ages of 55 and 74 years, have a 30 pack-year history, and have not quit smoking within the last 15 years are encouraged to undergo lung cancer screening, according to Inga T. Lennes, MD, MBA, MPH.
Screening among this patient population has significant morbidity benefits, as seen in the NELSON trial. In this study, more than 15,000 patients were enrolled, half of whom received CT screenings at baseline and years 1, 3, and 5.5, while the other half of patients did not receive screenings. The CT scan protocol included centralized reading of the images as well as monitoring of lung nodule volume and volume doubling time.
Ten-year follow-up results showed that CT scans decreased mortality by 26% in asymptomatic men who were at high risk for lung cancer. Additionally, in a subset of women who underwent screening, it was found that the risk of dying of lung cancer was reduced by 39% to 61% in various years of follow-up. According to Lennes, further subgroup analyses are studying the reasons why women benefitted so greatly from screening.
“One of the fastest growing populations of patients who are being diagnosed with lung cancer are never-smoking females. They aren’t the largest population that is being diagnosed with cancer, but it is definitely growing over the years,” said Lennes, medical director of Ambulatory Services and director of Clinical Quality at the Massachusetts General Hospital Cancer Center. “Finding out what the drivers are of lung cancer in women is very interesting work and we're very excited to learn more about that.”
Moreover, efforts are continuously being made across institutions to better implement these screening protocols.
In an interview with OncLive, Lennes discussed the recommended screening protocol for patients at risk for lung cancer, maintaining patients post-screening, and how physicians can best assist this patient population.
OncLive: What is the standard protocol and recommendation regarding lung cancer screening?
Lennes: As of 2013, the recommendation is for patients in the United States who are between the ages of 55 to 74 who have a 30 pack-year history of smoking, which means they've smoked at least 1 pack of cigarettes a day for 30 years, or the equivalent, such as smoking 2 packs a day for 15 years. If you've smoked that equivalent of cigarettes and you are either a current smoker or you have not quit within the last 15 years, you're eligible for a lung cancer screening.
There are some other considerations. Your primary care doctor, or whoever refers you to screening, needs to do a shared decision-making visit with you where you discuss the pros and cons of screening. If patients have several comorbidities or are otherwise very ill, sometimes we forgo screening because of those considerations. However, the major considerations for screening are age, smoking status, and [whether or not they have quit screening].
Are patients who fall short of that criteria ineligible for screening?
For most patients, the screening wouldn't be covered by insurance; however, there are other organizations that recommend slightly different screening criteria. One of those organizations, for example, is the National Comprehensive Cancer Network; they have slightly different recommendations for screening. They're a little broader and incorporate patients who have a lower pack-year smoking history and a broader age range.
Some data show that broader-age categories and more diverse categories of risk may also benefit from screening. We saw that in the NELSON trial, which had patients in a broader age range and lower pack-year history of smoking go through screening. Researchers found these patients also derived benefit from screening. What we will see in the future is further refinement of the risk categories that patients fall into, followed by tailored screening recommendations for those risk categories.
Could you elaborate on the NELSON trial findings and the gender disparities?
The NELSON trial confirmed the finding that were first established by the National Lung Screening Trial (NLST), showing a lung cancer screening mortality benefit. The mortality benefit in the NELSON trial exceeded the 20% mortality benefit we saw in NLST. For male patients who were part of the NELSON trial, the mortality benefit was 26%.
Strikingly, for women, the mortality benefit was greater than 60%. That subgroup showed [lung cancer screening] had a significant benefit in the NELSON trial. There is a further subgroup analysis in ongoing studies trying to identify exactly what the factors are in that population that would lead to that magnitude of benefit. We are very excited to see that.
There were roughly equal numbers of men and women in the NELSON trial. There may be some selection bias in terms of who is electing themselves to be part of a clinical trial—that's always something that we wonder about.
One of the other issues is that in screening, there are cancers that we find because someone has never had a CT scan before. That’s what we call interval cancers, or cancers that we find during the screening process. Some of those differences, in terms of the types of cancers we're seeing, may differ between the sexes. We'll have to look for those subgroup analyses in the future to know more.
Since these data have read out, have there been any difference in screening rates?
There have been some important studies that have shown that the rate of lung cancer screening has been sluggish and that we could do a better job of screening a larger population. I get the sense that momentum is changing since the publication of the NELSON trial.
In our own institution, the rate of screening has jumped dramatically over the past year. Part of that is due to the infrastructure necessary for managing screening findings. In the beginning, when screening was showing to improve mortality, there was a rush of screening programs that were set up appropriately. What we're finding now is that the infrastructure around a screening program is really important. Making sure that your subspecialists, not only in thoracic surgery but in radiation oncology and interventional radiology, who are doing ablations as well as stereotactic body radiation therapy and minimally invasive surgeries, are aligned around the screening programs is important. This is so that patients who have pulmonary nodules can be seen and cared for in a multidisciplinary manner.
Are protocols in place to ensure that patients keep coming to the clinic after screening?
One of the most important aspects of our clinic is our navigation personnel. We have an access nurse that serves as a navigator in the pulmonary nodule clinic, and we found that has improved adherence by 100%. We have other data, which show that patients adhere to guideline recommendations for follow-up CT scans about 40% to 50% of the time within our institution. By adding a nurse navigator, that adherence rate is well over 90% and approaching 100%. Adding the personnel that's necessary to ensure adherence is important.
Also, with these complex patients, we need someone to make sure they're being navigated and don't duplicate services. The navigator must make sure the patient is not seeing both pulmonologist in the nodule clinic and a pulmonologist outside of the nodule clinic, or a surgeon inside and outside. The access nurses are incredibly important.
What other ongoing research is being done related to lung cancer screening?
Some of the most exciting work that's being done in lung screening is to look at some of the new techniques for volume metric analysis, different artificial intelligence, and deep learning algorithms. The human eyes are only so good at being able to detect these small nodules. While our radiology colleagues are fantastic, the exciting part of lung cancer screening will be to see what our computers can bring to the recognition aspect of this in terms of being able to look through large volumes of imaging data and to find cancers.
I have a friend who works at Massachusetts Institute of Technology and is focusing on deep learning and models in radiology and screening. What I'm told is that the future of finding cancer is going to be in the data because the immense amount of data we get from images has only been processed by the human eye. Once we apply the metrics of artificial intelligence to those data, our computers will help us find cancer in the data we already have.
What advice can you give to your colleagues regarding the importance of screening?
The first piece of advice is to not assume that patients aren't eligible for screening. I talk with a lot of primary care physicians who may assume that because of certain comorbidities, a patient could never be eligible for a surgery. What that provider may not realize is that there are minimally invasive surgeries and robotic techniques that a patient may be eligible for.
There is also radiation. We can almost always offer radiation to patients who aren't eligible for surgery. Now there are new techniques, such as radiofrequency and cryoablation, which offer us a lot of options for patients who have had radiation in the past or other things that would make them ineligible for other modalities. Finding a multidisciplinary clinic to refer patients who have a suspected early-stage cancer or a pulmonary nodule finding on screening is important so they can get a comprehensive review and recommendation and they don't miss out on potentially curative therapy.
Is there anything being done at your institution that you would like to discuss?
The most exciting work in this area that we're doing is trying to marry biomarker analysis with the lung cancer screening and pulmonary nodule clinics. With every patient that comes to our clinic, we have the ability to offer them enrollment in some of our biomarker collection protocols. We are collecting blood samples and pairing that with their data and their clinical outcomes to understand more about the evolution of early cancer—so that we can catch it before it starts.
De Koning H, Van Der Aalst C, Ten Haaf K, et al. Effects of volume CT lung cancer screening: mortality results of the NELSON randomized-controlled population based trial. J Thor Oncol. 2018;S185. doi: 10.1016/j.jtho.2018.08.012