November 12, 2024 • 36 mins
In this episode, we speak with Dr. Natalie Lui, Assistant Professor of Cardiothoracic Surgery at Stanford University, about lung cancer screening and the importance of early detection. Since lung cancer is often diagnosed in advanced stages, yearly low-dose computed tomography (LDCT) screening is crucial, especially for heavy smokers. We discuss risk factors, screening criteria from the U.S. Preventive Services Task Force, National Comprehensive Cancer Network, and the American Cancer Society, and why screening isn't universal despite its importance. Additionally, we explore current research, recent updates to guidelines, barriers to screening in underserved communities, and the potential of emerging technologies, such as AI, to enhance lung cancer screening in the future.
CME Information: https://stanford.cloud-cme.com/medcastepisode92
Claim CE: https://stanford.cloud-cme.com/Form.aspx?FormID=3122
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
(00:00):
(bright music)
(00:06):
- Welcome to Stanford Medcast,
the podcast from Stanford CME
that brings you the latest insights
from the world's leadingphysicians and scientists.
If you're joining us for the first time,
be sure to subscribe on Apple Podcasts,
Amazon Music, Spotify, or YouTube
to stay updated with our newest episodes.
(00:26):
I am your host, Dr. Ruth Adewuya.
Joining me today is Dr. Natalie Lui.
Dr. Lui is a boardcertified thoracic surgeon
and assistant professorof thoracic surgery
at Stanford School of Medicine.
After earning her MD
at Johns Hopkins School of Medicine,
Dr. Lui completed aresidency in general surgery
(00:48):
at the University ofCalifornia San Francisco,
where she became involved inthoracic oncology research
and completed a Master of Advanced Studies
in Clinical Research.
She went on to hold afellowship in thoracic surgery
at Massachusetts General Hospital
and continues to specializein thoracic oncology.
(01:09):
Dr. Lui's clinical focusspans thoracic surgery
with special attention to lung cancers
and the use of robotics in surgery.
She additionally conducts research
related to increasing ratesof lung cancer screening
using intraoperative molecular imaging
for lung cancer localizationand robotic surgery education.
(01:32):
Dr. Lui, thank you so muchfor chatting with me today.
- Thank you so much for having me here.
- I'm excited to talk toyou about lung cancer,
but specifically, lung cancer screening.
There's a lot we cancover about this topic,
but I thought a reallygreat place for us to start
is at the high levels.
Can you provide a really highlevel overview of lung cancer,
(01:54):
some of the underlying pathology
that will set the stagefor our conversation?
- Lung cancer starts astumors that grow in our lungs,
which are the two main organs
that sit in our chestand provide oxygen to us.
And there are several different kinds.
Sometimes people will referto small-cell lung cancer
(02:15):
or non-small cell lung cancer.
The most common type is a typeof non-small cell lung cancer
called adenocarcinoma.
And lung cancer is the most common cause
of cancer-related death inthe United States, worldwide.
It's not really that well known,
so it is the second mostcommon type of cancer
(02:39):
in women after breast cancer,
and the second most common type of cancer
in men after prostate cancer,but it is much more deadly
than either of those types of cancers,
and it causes the mostnumber of cancer deaths,
again, in the United States and worldwide.
- I think that's important to note,
because we talk a lot aboutbreast cancer in women,
(03:03):
we talk about prostate cancerin men, but that statistic,
the most common cause of cancer deaths.
- That's right, in the US,
about 20% of people who die from cancer
die from lung cancer,
so this is about expected 125,000 people
in the US this year.
(03:23):
- And I'm sure we'll diveinto it a little bit more,
because the one thingthat comes to mind for me
is the equation of lungcancer with smoking,
and given that number, itcan't all be about smoking.
And I don't wanna jump ahead of myself,
that's an importantpoint of a conversation.
But let's talk aboutlung cancer screening,
(03:46):
because that is such a hugenumber of cancer deaths.
Lung cancer screening is so important.
What does lung cancerscreening typically involve
and why is early detection so critical
in improving patient outcomes?
- Someone could have asmall tumor in their lung
and feel completely normal.
(04:06):
They could breathe normally.
They could go about their day
and exercise and they have no idea.
It's usually only when a lung cancer
becomes a lot bigger or spreads
that it would cause any symptoms.
So for example, if a tumor got so big
that it was pressing on an airway,
someone might start coughingor coughing up blood.
(04:28):
If the tumor spread to the brain,
someone might start having headaches,
or if it spread to the spine,
someone might start havingback aches or back pain.
And so, it's only at thatpoint when people have symptoms
that they would usually go to their doctor
and get evaluated.
And you can see these symptomsare all pretty common,
(04:49):
cough, headache, back pain,
so even when they go to their doctor,
it might take a while to getthe diagnosis of lung cancer.
So in the US, most people whoget diagnosed with lung cancer
have advanced disease.
We have a stagingsystem, one through four,
and one is the earliest, fourmeans that it has metastasized
(05:11):
or spread to another organ,
and most people are diagnosedin stage three or four,
because that's when they have the symptoms
and go get evaluated.
So, screening means looking for people
before they have symptoms.
And now, we currentlyscreen using a low-dose,
computed tomography scan or CT scan
(05:32):
or CAT scan is another namewe have for it, so LDCT,
and for heavy smokers in the US,
we recommend an annual LDCT.
- I think that the fact that the symptoms
are things that you could attribute
to a super active weekend,
maybe you strainedyourself and had back pain,
(05:55):
probably adds to thechallenge of early diagnosis.
I'm curious about how do you delineate
between, oh, it's just really back pain,
and oh yeah, I just havethe common cold and cough.
How do you handle that?
- That's one of the hard parts
about being a primary care physician.
If someone comes in with a cough,
they probably don't have lung cancer,
(06:17):
but it is important togo try to figure out
what is causing the cough and treat it,
and if it doesn't go away,
then to become morerestive in the evaluation.
And so a lot of times,
people will end up gettinga CT scan for a cough.
This is a diagnostic CT scan,
because they're responding to symptoms,
(06:39):
instead of the low-dose CT scan,
which is what we recommend for screening,
which is before someone has any symptoms.
- Smoking, we all know, is arisk factor for lung cancer.
How does screening benefit smokers
and what other groups shouldclinicians consider screening?
(06:59):
- Smoking is definitelythe primary risk factor
for lung cancer.
In the US, about 80% of peoplediagnosed with lung cancer
have smoked at some point in their lives.
And so, the currentguidelines for screening
only recommend screeningfor heavy smokers.
So there are three criteria,age, 50 to 80 years old,
(07:23):
a smoking history ofat least 20 pack-years,
and either a current smokeror quit within 15 years.
Now, a lot of people haveasked me what a pack-years is.
We use this commonly in medicine
to quantify what someone'ssmoking history has done.
And so to estimate theirexposure to cigarette smoking,
(07:45):
we multiply the averagenumber of packs per day
that they smoked by the numberof years that they've smoked.
So, one example of someone
who would meet a 20-pack-yearthreshold for screening
would be someone who smoked an average
of one pack per day for 20 years,
and you multiply those and you get to 20.
(08:07):
But there are a few differentways you can get to 20.
So if someone is a really heavysmoker at two packs per day,
they would only haveto smoke for 10 years.
- So it's helpful to understand
that there is a specific criteria
that clinicians should be aware of
when considering screening for smokers.
What other populationsshould be considered?
(08:29):
- Right now, the USPreventive Services Task Force
or USPSTF, those are thecriteria that they use,
and that is what is covered byMedicare and other insurance.
But we do know thatother people are at risk.
There are other societies thathave different guidelines.
So for example, the NationalComprehensive Cancer Network,
(08:51):
or NCCN, has a secondgroup of risk factors
that they consider, likeexposure to secondhand smoke,
exposure to occupational chemicals, radon,
lung diseases, like tuberculosis or COPD,
a family history of lung cancer,
those are all part oftheir list of risk factors,
(09:14):
but it's currently notcovered in the United States.
And then the American CancerSociety as another example
of how there is some disagreement
about what the criteria should be,
the American CancerSociety recently removed
the quit date criteriafrom their recommendations,
and so they only recommendthe age of 50 to 80
(09:36):
and the pack-years of 20.
And they thought that it wouldbe probably more equitable
to get rid of that and easier,
'cause it's just a lotfor people to remember
to figure out whetherthey're eligible or not.
- I can see how multiplerecommendations can be challenging,
but then ultimately,
it's who's gonna payfor the screening test.
(09:58):
What are your thoughtson the idea of screening
being recommended for all patients
regardless of their risk factors?
- That is a really good question.
So, the low-dose CT scans do have risks.
Besides the cost of doing the scan,
there's radiation exposure.
It's about the same amount as six months
(10:18):
of natural background radiation,
but it's not something that we want to do
for someone with verylow risk of lung cancer.
Other risks include false positives,
so finding a lung nodulethat's initially worrisome,
that might prompt additionaltests or even biopsies,
sometimes even surgery to remove it,
(10:39):
which end up being benign or negative,
and so that's a lot thatsomeone might go through.
There's a lot of anxietywith the tests as well.
And then, there's also a riskthat we would find things
that are abnormal thatwouldn't hurt somebody
that we end up treating.
And so because of these risks,
we don't recommend lung cancerscreening for everybody,
(11:03):
but there is some thoughtthat people outside
of the current criteriacould probably benefit,
and we need to do some more research
to figure out which groups those are.
- It sounds like there'sa potential to expand
but not expand to all,but more data is required.
- That's right.
If we screened everybodystarting at a young age,
(11:24):
we would find all lung cancers early
if that was somehow possible.
But besides our healthcare system
not being able to handle thatmuch screening and testing,
I think it probablywouldn't be a good idea
for people who are really at little risk.
It's hard because we do have outliers.
We see people who are very young
who wouldn't meet screening criteria,
(11:46):
and we've seen them diagnosedwith advanced lung cancer,
and it's hard not to wonder who else,
how could they have been screened?
Who else should be screened
to prevent these cases from happening?
- Is there a higherrisk around lung cancer
when you talk about Blackpopulations, Latino, Asians
or is that an independent variable?
- One of the biomedicalinformatics faculty here,
(12:09):
Summer Han, as well as one of our former
Stanford thoracic surgery fellows
who's now at MassachusettsGeneral Hospital, Jeff Yang,
both of them have done researchon the screening criteria
and how that affects disparitiesbetween African-Americans
and other racial populations in the US.
And so interestingly,
(12:30):
they found that using thenumber of years smoked
instead of pack-yearswould decrease disparities.
So they look at thingslike what percentage
of people in differentgroups with lung cancer
would have qualified for screening?
And so when they use the current criteria,
(12:50):
the proportion of AfricanAmericans who would qualify
is much lower than the proportionof Caucasian Americans.
So when they trieddifferent screening criteria
and used, for example,the number of years smoked
instead of pack-yearsand found that actually
a lot more of those AfricanAmerican people with lung cancer
would have qualified in that way.
(13:12):
As you can imagine, thisresearch is pretty difficult,
because there's one a set of research
where it looks at people withcancer, and so you can see
whether they would've met thescreening criteria or not.
But what even better is if youhave the entire denominator,
the cohort of healthy people
and the people who willdevelop lung cancer
(13:32):
and then see how many peoplewould you have to screen
in order to find a certain number
of people with lung cancer.
And so, Dr. Yang had looked at databases
in the southern United States.
Dr. Han has looked at databasesin California and Hawaii,
of course, very differentracial and ethnic makeups there.
But they have found similar things
(13:52):
that changing the criteria from pack-years
to years of smoking mightactually decrease the disparities
a lot among different groups.- That's incredible.
So, what do we do with this information?
What's the next step?
Because it sounds like theguidelines need to change.
- I know, I think it's really hard,
because it is hard to get this data,
(14:14):
and none of this dataaddresses the mortality benefit
in the (indistinct) atthis randomized trial
of 50,000 people did.
And that's not feasible to repeat that
for every group thatyou are interested in.
And so I'm very gladthat medical informatic
and statistician like Dr. Hanare looking at these databases
(14:36):
to try to extract the information
without doing a huge trial.
And I do think that,
as we continue withthese research avenues,
that we may be able to change guidelines.
It's slow but necessary work.
Another group that's affected
are Asians and Asian Americans.
We talked about never smokerswho develop lung cancer.
(14:57):
In the US, about 20% of peoplediagnosed with lung cancer
had never smoked.
And in Asia and amonggroups of Asian Americans,
it ranges all the way up to80%, it's very different.
It seems like a verydifferent type of disease.
And so because of that, thereare many countries in Asia
(15:19):
who are offering screening already
even to people who have never smoked.
So these guidelines are very different
than what we find in the US.
And one of the most developedstudies was in Taiwan,
called the TALENT study,and they enrolled people,
men and women who had never smoked,
who had one other risk factor,
so that could be secondhandsmoke or family history,
(15:42):
and they actually found thatthey detected lung cancer
in 2.1% of that group,which is even higher
than what was found in the heavy smokers
in the national lung screening trial.
And again, about half ofthe people had stage one,
so this was definitely findingmore early-stage disease.
(16:02):
And based on that study,
they started the Early Detection Program,
so they're currently screeningpeople with family history,
so never smokers with family history,
because they found that wasreally the biggest risk factor
for the people who had never smoked.
- That's a really greatsegue to my next question.
You have a really great example of Taiwan
(16:23):
and what we can learnfrom these approaches.
I completely understand thatthe United States is unique.
We are made up of a diverse population,
and so it's not a one-size-fits-all,
which it's part of its beauty,but it's also challenging
to address some of these health issues.
Are there any other countries
that have taken that similar approach
(16:45):
that have been successful or not?
- There are several countries in Asia
who have been screening never smokers,
but I think the largest one is in Taiwan,
easiest for us to lookat their publication.
It's been fascinatinglearning about screening
around the world.
There's a world conferenceon lung cancer every year
that the International Association
for the Study of Lung Cancer puts on,
(17:07):
and so they always havepeople from around the world
giving talks about screeningprograms in their countries
and what they're doing to implement them.
And it's really a very wide range
from countries who don't haveany screening programs at all
to ones that are just completely covered.
And so, I think the US, you know,
we know that we have a long way to go,
(17:27):
and we have ways tolearn from other places
in terms of implementation
and getting people who areeligible in for screening.
- We talked about low-dosecomputed tomography, LDCT.
Can you just give us alittle bit more insight?
How does it work and what has made it,
it sounds like the preferred method,
for lung cancer screening?
(17:49):
- Computing tomographyor CT scans, in general,
are a mix of X-rays andthen computer assistance.
And so essentially, it putstogether a very detailed picture
of the lungs and really the entire chest,
where we can see really tiny lung nodules
using these CT scans.
(18:10):
And the technology has gotten really good
so that the radiation exposure for LDCT
has become quite low, low enough
that we feel comfortablerecommending an annual scan
for people for screening.
I will add the low-doseCTs are really quick,
so they just take a few minutes.
They don't hurt at all.
(18:30):
People just have to climb onto the bed
and then they get moved into the scanner,
which is like a donut.
And for the LDCTs, there'sno needle involved,
no IV, no contrast, soit really is quite quick
and convenient for people.
- What were we doing before LDCTs?
- We were not doing any screening.
- Oh.- Yeah,
(18:51):
there were a lot of peoplelooking into chest X-rays,
which is one single film,rather than the LDCT,
which is many films in a lot more detail.
Some people looked into sputum samples,
so looking to see ifthere were tumor cells
in sputum that people would cough up.
But it wasn't until theNational Lung Screening Trial,
(19:12):
which was performed in theUS, then published in 2011,
that showed that therewas a mortality benefit.
So that trial, the NLST was sponsored
by the National Cancer Institute,
and it was a multicentertrial across the US
that enrolled over 50,000 people
and randomized them to eitherthe LDCT or the chest X-ray.
(19:37):
These were all heavy smokers,and they found that the group
that got the annual LDCT
had a 20% lower deathrate from cancer, 20%,
and so that's reallywhat sparked the change
in the guidelines in the US.
So that paper was published
in "The New England Journalof Medicine" in 2011.
(19:57):
The USPSTF recommended screening
based on those criteriaof the trial in 2013,
and then Medicare startedcovering it in 2015.
And since then, we'vemade some smaller changes
in the eligibility criteria.
There were some studies inEurope that showed that people
(20:19):
with slightly lower smokinghistory might benefit.
And so the USPSTF has revisedtheir criteria a few years ago
to take into account that new data.
- Well that's interesting,2015 is not long ago,
right, it's pretty new?
I am amazed by how innovative
(20:39):
and advanced medicine and clinical care is
and how it can be in so many areas,
but also, how it is not in so many areas.
- It's really fascinating tosee this change over time,
because I was essentially in training
and a new surgeon whenall of this was happening.
(21:01):
And so to see it in real timeand see how health policy
then reflects the medical research
has been really fascinating.
And a lot of people ask why surgeons
are involved with screening,
because it's really seenas a primary care arena,
and I think surgeons really saw the impact
that it made on patients,because I can operate
(21:22):
on someone with stage-one disease
and potentially offer them a cure.
By the time it gets tostage four and it's dense,
it's very rare that it can offer anything.
And so we love seeing screening,
and we're huge advocates for it,
because we want morepeople to be diagnosed
in that early stagewhen we can be helpful.
- What an incredible outlook too.
(21:44):
You've talked about the recent changes
in the screening guidelines.
I wanna pivot to some of thebarriers that patients face
when accessing lung cancer screening.
What have you seen, especiallyin underserved populations?
- That is a really great question,
because despite lung cancer being covered
(22:06):
for almost a decade now, ourscreening rates are really low.
In California, it'sestimated that only 16%
of eligible people are getting screened.
That's a lot of people that might benefit
that are not getting their annual screens.
We've looked into many of these barriers,
and I think that it isreally multifactorial.
(22:28):
I do feel like a lot of peoplestill don't know about it.
It's been almost a decade,but it's still really new.
We've done some surveysat community health fairs
and found that the percentage of people
that know about lung cancer screening
is far below people who know
about mammograms for breast cancer
or colonoscopies for colon cancer.
(22:49):
It is still just not widelyknown in the community.
I think another part isthat primary care physicians
are usually this firstaccess point to screening,
and they have been really overwhelmed.
It's hard to fit screening,
which is, by definition, noturgent, into a short visit
when you have many othermedical problems to discuss,
(23:11):
and so that's a big part of it.
And I think making it muchmore part of the workflow
for primary care physicians
and assisting them andsupporting them in that
is really important.
And then there are different factors
in terms of the health's policy,
which may be larger andwe could really change,
but it's pretty confusing toknow how much people will pay.
(23:32):
It all depends on their insurance.
Things are covered,but there are questions
about copay depending on insurance,
and I think a lot of that
just reflects our healthcare system,
and it's hard for us to answersome of those questions.
- I think you're spot on there.
I can see how all of thosefactors would impact.
I'm curious if you've also seen
(23:52):
whether socioeconomic factors,whether it has an impact
to adhering to lung cancerscreening guidelines.
Because I think what I heard from you
is that these are annual, right,
so it's not just a one and done?
- I don't know specificallyabout adherence,
but there have been a lot of studies
about disparities among groups,
(24:13):
racial and ethnic groups interms of lung cancer screening.
When you look back at theNational Lung Screening Trial,
which prompted all of this,
over 90% of the participants were White.
So about, I think less than5% were African American,
about 2% were Asian, soit doesn't totally reflect
(24:33):
the US makeup and even worldwide,then it really does not.
And then the studies that followed
were mostly done in Europe,
so they're really very differentin terms of racial makeup.
And one of the biggest onescalled the Nelson Trial,
they actually had about 85% men,
(24:54):
so women were a sub-analysis in the group.
So, these trials arehuge, and they gave us
a lot of information aboutthe mortality benefit,
but they definitely didn't answer
all of the questions that we had
in terms of who wouldbenefit from screening.
I think that it isreally hard to make sure
(25:14):
that people get screeningand know about it,
and then also, to adhereto the annual scan.
A lot of it is very easy,like the scan is very quick,
but it does requiretransportation to go get it,
if there are language barriers
in terms of understandingthe eligibility or adherence,
the time out of the day to go there,
(25:36):
so there are a lot of nuances that I think
would make adherence andfollow up a lot better.
- I think what you saidjust reflects the challenges
that exist in clinicaltrials and in general,
and not just for lung cancer,the need for more diversity
and more representationin clinical trials.
(25:58):
I think there's a growing recognition
of the need for that inthe scientific community.
I wonder whether psychological barriers
can also impact the patient
in terms of wanting togo through screening
where it's almost this notion
of ignorance is bliss type thing.
(26:19):
What are your thoughts on that?
- I think you are totally right.
I think it is really scary tothink about having lung cancer
and having a test that might find that
and really just change your entire life.
I think it's also hard,
because screening isbased on smoking history,
and we've known thatsmoking as a risk factor
(26:40):
for lung cancer for a decade now.
And so I think there is some stigma
in terms of being a smoker,
smoking despite knowingthat there is that risk,
and then potentiallyfeeling like you caused this
if you were smoking and have lung cancer.
Now a lot of people whosmoke don't get lung cancer,
(27:02):
and there are a lot ofpeople who have never smoked
who develop lung cancer.
So I try to reduce thatand really get people
to feel more open to screening,
but I can see that it is hard.
It's a very scary prospect.
- You alluded to this earlier
when we were talking about LDCT,
the false positive that can come up,
(27:23):
and I think it's one of the things
that it has been criticized for,
it's high false positive rates,potential over diagnosis.
How do clinicians balance this concern
when there are clearbenefits for early detection?
- Yeah, different clinicians
will respond differently to this question.
You're talking to one of the leads
(27:44):
of the lung cancer screeningprogram at Stanford,
but if you look at theNational Lung Screening Trial,
they quote a really high falsepositive rate of almost 25%,
which seems pretty ridiculous.
When you look at the details though,
they count even a nodulethat's four millimeters
as a positive, and so theseare nodules that are so small
(28:07):
that we wouldn't recommenddoing anything else
based on that nodule alone.
We would probably recommendanother scan in a year,
which is what is recommended anyway.
So it would still count as a positive
under the NLST definition,but wouldn't actually prompt
any invasive evaluation at that point.
(28:27):
So if you look at the falsepositive rate in NLST,
for the first couple years,it was around just under 25%,
and then for the third year,
suddenly that falsepositive rate drops to 7%,
and that's because if a nodulewas followed and didn't grow,
then it would be called positive initially
(28:48):
and then switched to negative.
And so it's interesting, Ithink that some of the numbers
of a high false positive rate
don't really reflect the risk to patients,
which is really the risk
when we decide to do more evaluation,
so invasive biopsies or testing
that might actually havecomplications that affect it.
(29:08):
- I think what I'm hearing you say,
and correct me if I'm wrong,
what I'm hearing you say is,"Do it anyway," and basically,
"Do a repeat LDCT before maybe taking
"any invasive measures."
- It depends on what is found.
American College ofRadiology actually came up
with a whole classificationsystem of lung nodules,
(29:31):
just to make sure thatthings were standardized,
and we could talk aboutthis false positive
or false negative rate.
They call it Lung-RADS,
lung and then reportingand data system, Lung-RADS,
and it goes from essentially zero to four,
and so each nodule that's found
is classified as one of these Lung-RADS.
(29:52):
And for each Lung-RADS,
they have an expectedpercentage that are malignant.
They expected prevalencewithin a population
and then the recommendations.
For example, if you have a Lung-RADS 2,
which is probably benign,
they would say get the scan in a year,
which is what someone would doif they didn't find a nodule.
(30:14):
And so, a lot of those small nodules
on NLST that were positiveactually would just go along
with this Lung-RADS classification,
and nothing would actuallybe different for that person.
Now as you get higher in Lung-RADS,
you get to ones that are very suspicious,
and over 50% of those willbe diagnosed as lung cancers.
(30:37):
And those are the onesthat you'd recommend,
they call it tissue sampling or biopsy,
and that can even involve surgery.
And so I think that is when
we have a much more detaileddiscussion with a patient
about the options and the risks
for these invasive procedures.
- That's very helpful to clarify.
I'm going to revise what I'mhearing from you at this point.
(30:59):
Do it anyway, but your analysis
of this quote-unquote false positive
is a little bit more nuanced,
and there's additional classifications
that you need to utilize todetermine what to do with that.
- The vast majority of peoplewho have lung nodules found
(31:19):
many times what we dofor those lung nodules,
especially when they'rereally small, is just watch.
When we find really small lung nodules,
they're too small tobiopsy or test in any way,
and so the recommendationis just to follow them.
So in that sense, if thatcounts as a false positive,
(31:39):
that's something that won'tnecessarily harm someone.
I think where we're worriedabout finding a false positive
is when they undergo procedures,
like a bronchoscopy, abiopsy, again, even surgery.
Sometimes I will remove a nodule
that looks really suspicious with surgery,
because I'm so worried about it.
(32:01):
And we celebrate when it'snothing, but in that sense,
that patient didn't need it,we just didn't know for sure.
It is fair that we don'twant to expand screening
to everybody, 20-year-old, veryunlikely to have lung cancer
and any nodule found is muchmore likely to be benign.
So in that sense, screening someone
(32:22):
who's clearly at extremely lowrisk doesn't make any sense,
because there'd be a lot of anxiety,
potential other procedures or tests
when most likely, not lung cancer.
The question is where to draw that line,
and I think that atleast for heavy smokers,
it's been answered.
The NLST, despite thesefalse positive rates,
(32:44):
showed a mortality benefit,
so that is an easy,"Yes, go get screened."
And the other question is other groups,
where we think they're at higher risk,
but it's a little bit more unclear
what that mortality benefit would be.
That's where it's a little bit harder,
and we need more research.
- Are there any emerging screening methods
or technologies that could either enhance
(33:07):
or complement some of the LDCTs
in our current lung cancerscreening practices?
- I think there are a couple avenues.
One is that the technology for CT scans
keeps getting better, andso some of our radiologists,
Henry Guo is one of them here at Stanford,
is looking into photon counting CT scans
(33:29):
and whether they couldadminister screening tests
with even lower dose CT,
so we called that ultra-low dose CT scan.
And then blood-based tests area huge upcoming technology.
We had one of our radiationoncologist, Max Diehn here,
who's been doing research intesting, circulating tumor DNA
(33:52):
in blood sample for many years.
And I think those will beamazing when we could do,
that'll be even betterthan getting a CT scan
if we could do that for lung cancer
and potentially other cancers.
- That really sounds exciting.
In the past few years at least,
AI is everything everyone's talking about.
Has that been in the lungcancer screening space as well?
(34:15):
Have we been talkingabout applications for AI
in lung cancer screening?- Yes, for sure.
There have been a couple different ways
people have started looking at AI,
but I think that there is alot more to go in the future.
Reading the CT scan, Ithink, is one big part,
so finding lung nodules, we'refollowing them over time.
It's something that takes a lot of skill
(34:37):
from our radiologists andso trying to figure out
if AI can be helpfulin that would be great.
And then I think also looking at risk
for lung cancer in general
would be another potential application.
None of these are beingused right now consistently,
so I think, again, we have alot more to learn about them,
but I'm very excited aboutpotential applications.
(35:00):
- Looking forward, what do you envision
as the future of lung cancer screening?
- That is a really good question.
I am hoping that one,
we get people who weknow are eligible now,
we know there's a mortality benefit,
that we find a way to reachthem and get them screened.
I think that's the most important thing.
(35:20):
And then second, we know thatother people will benefit,
and we're just not sure who exactly,
and so I think that researchwill be really important
to figure out how else we cansave lives with screening.
And then the last thing,which you already asked about
is other methods of screening.
I think in the future,these blood-based tests,
(35:43):
any tests that will make it easier
for people to get screened and diagnosed
will be better and really help save lives.
- This has been a wonderful conversation
with so many key takeaways for clinicians,
whether you're a primary care clinician
or a lung cancer specialist.
I so appreciate your time today.
- Thank you so much, Dr.Adewuya. It's been great.
(36:03):
(gentle music)
- This episode was broughtto you by Stanford CME.
To claim CME forlistening to this episode,
click on the Claim CME link below
or visit Medcast.stanford.edu.
Check back for new episodes
by subscribing to Stanford Medcast
wherever you listen to podcasts.
(36:24):
(bright music)
(bright music)
(00:06):
- Welcome to Stanford Medcast,
the podcast from Stanford CME
that brings you the latest insights
from the world's leadingphysicians and scientists.
If you're joining us for the first time,
be sure to subscribe on Apple Podcasts,
Amazon Music, Spotify, or YouTube
to stay updated with our newest episodes.
(00:26):
I am your host, Dr. Ruth Adewuya.
Joining me today is Dr. Natalie Lui.
Dr. Lui is a boardcertified thoracic surgeon
and assistant professorof thoracic surgery
at Stanford School of Medicine.
After earning her MD
at Johns Hopkins School of Medicine,
Dr. Lui completed aresidency in general surgery
(00:48):
at the University ofCalifornia San Francisco,
where she became involved inthoracic oncology research
and completed a Master of Advanced Studies
in Clinical Research.
She went on to hold afellowship in thoracic surgery
at Massachusetts General Hospital
and continues to specializein thoracic oncology.
(01:09):
Dr. Lui's clinical focusspans thoracic surgery
with special attention to lung cancers
and the use of robotics in surgery.
She additionally conducts research
related to increasing ratesof lung cancer screening
using intraoperative molecular imaging
for lung cancer localizationand robotic surgery education.
(01:32):
Dr. Lui, thank you so muchfor chatting with me today.
- Thank you so much for having me here.
- I'm excited to talk toyou about lung cancer,
but specifically, lung cancer screening.
There's a lot we cancover about this topic,
but I thought a reallygreat place for us to start
is at the high levels.
Can you provide a really highlevel overview of lung cancer,
(01:54):
some of the underlying pathology
that will set the stagefor our conversation?
- Lung cancer starts astumors that grow in our lungs,
which are the two main organs
that sit in our chestand provide oxygen to us.
And there are several different kinds.
Sometimes people will referto small-cell lung cancer
(02:15):
or non-small cell lung cancer.
The most common type is a typeof non-small cell lung cancer
called adenocarcinoma.
And lung cancer is the most common cause
of cancer-related death inthe United States, worldwide.
It's not really that well known,
so it is the second mostcommon type of cancer
(02:39):
in women after breast cancer,
and the second most common type of cancer
in men after prostate cancer,but it is much more deadly
than either of those types of cancers,
and it causes the mostnumber of cancer deaths,
again, in the United States and worldwide.
- I think that's important to note,
because we talk a lot aboutbreast cancer in women,
(03:03):
we talk about prostate cancerin men, but that statistic,
the most common cause of cancer deaths.
- That's right, in the US,
about 20% of people who die from cancer
die from lung cancer,
so this is about expected 125,000 people
in the US this year.
(03:23):
- And I'm sure we'll diveinto it a little bit more,
because the one thingthat comes to mind for me
is the equation of lungcancer with smoking,
and given that number, itcan't all be about smoking.
And I don't wanna jump ahead of myself,
that's an importantpoint of a conversation.
But let's talk aboutlung cancer screening,
(03:46):
because that is such a hugenumber of cancer deaths.
Lung cancer screening is so important.
What does lung cancerscreening typically involve
and why is early detection so critical
in improving patient outcomes?
- Someone could have asmall tumor in their lung
and feel completely normal.
(04:06):
They could breathe normally.
They could go about their day
and exercise and they have no idea.
It's usually only when a lung cancer
becomes a lot bigger or spreads
that it would cause any symptoms.
So for example, if a tumor got so big
that it was pressing on an airway,
someone might start coughingor coughing up blood.
(04:28):
If the tumor spread to the brain,
someone might start having headaches,
or if it spread to the spine,
someone might start havingback aches or back pain.
And so, it's only at thatpoint when people have symptoms
that they would usually go to their doctor
and get evaluated.
And you can see these symptomsare all pretty common,
(04:49):
cough, headache, back pain,
so even when they go to their doctor,
it might take a while to getthe diagnosis of lung cancer.
So in the US, most people whoget diagnosed with lung cancer
have advanced disease.
We have a stagingsystem, one through four,
and one is the earliest, fourmeans that it has metastasized
(05:11):
or spread to another organ,
and most people are diagnosedin stage three or four,
because that's when they have the symptoms
and go get evaluated.
So, screening means looking for people
before they have symptoms.
And now, we currentlyscreen using a low-dose,
computed tomography scan or CT scan
(05:32):
or CAT scan is another namewe have for it, so LDCT,
and for heavy smokers in the US,
we recommend an annual LDCT.
- I think that the fact that the symptoms
are things that you could attribute
to a super active weekend,
maybe you strainedyourself and had back pain,
(05:55):
probably adds to thechallenge of early diagnosis.
I'm curious about how do you delineate
between, oh, it's just really back pain,
and oh yeah, I just havethe common cold and cough.
How do you handle that?
- That's one of the hard parts
about being a primary care physician.
If someone comes in with a cough,
they probably don't have lung cancer,
(06:17):
but it is important togo try to figure out
what is causing the cough and treat it,
and if it doesn't go away,
then to become morerestive in the evaluation.
And so a lot of times,
people will end up gettinga CT scan for a cough.
This is a diagnostic CT scan,
because they're responding to symptoms,
(06:39):
instead of the low-dose CT scan,
which is what we recommend for screening,
which is before someone has any symptoms.
- Smoking, we all know, is arisk factor for lung cancer.
How does screening benefit smokers
and what other groups shouldclinicians consider screening?
(06:59):
- Smoking is definitelythe primary risk factor
for lung cancer.
In the US, about 80% of peoplediagnosed with lung cancer
have smoked at some point in their lives.
And so, the currentguidelines for screening
only recommend screeningfor heavy smokers.
So there are three criteria,age, 50 to 80 years old,
(07:23):
a smoking history ofat least 20 pack-years,
and either a current smokeror quit within 15 years.
Now, a lot of people haveasked me what a pack-years is.
We use this commonly in medicine
to quantify what someone'ssmoking history has done.
And so to estimate theirexposure to cigarette smoking,
(07:45):
we multiply the averagenumber of packs per day
that they smoked by the numberof years that they've smoked.
So, one example of someone
who would meet a 20-pack-yearthreshold for screening
would be someone who smoked an average
of one pack per day for 20 years,
and you multiply those and you get to 20.
(08:07):
But there are a few differentways you can get to 20.
So if someone is a really heavysmoker at two packs per day,
they would only haveto smoke for 10 years.
- So it's helpful to understand
that there is a specific criteria
that clinicians should be aware of
when considering screening for smokers.
What other populationsshould be considered?
(08:29):
- Right now, the USPreventive Services Task Force
or USPSTF, those are thecriteria that they use,
and that is what is covered byMedicare and other insurance.
But we do know thatother people are at risk.
There are other societies thathave different guidelines.
So for example, the NationalComprehensive Cancer Network,
(08:51):
or NCCN, has a secondgroup of risk factors
that they consider, likeexposure to secondhand smoke,
exposure to occupational chemicals, radon,
lung diseases, like tuberculosis or COPD,
a family history of lung cancer,
those are all part oftheir list of risk factors,
(09:14):
but it's currently notcovered in the United States.
And then the American CancerSociety as another example
of how there is some disagreement
about what the criteria should be,
the American CancerSociety recently removed
the quit date criteriafrom their recommendations,
and so they only recommendthe age of 50 to 80
(09:36):
and the pack-years of 20.
And they thought that it wouldbe probably more equitable
to get rid of that and easier,
'cause it's just a lotfor people to remember
to figure out whetherthey're eligible or not.
- I can see how multiplerecommendations can be challenging,
but then ultimately,
it's who's gonna payfor the screening test.
(09:58):
What are your thoughtson the idea of screening
being recommended for all patients
regardless of their risk factors?
- That is a really good question.
So, the low-dose CT scans do have risks.
Besides the cost of doing the scan,
there's radiation exposure.
It's about the same amount as six months
(10:18):
of natural background radiation,
but it's not something that we want to do
for someone with verylow risk of lung cancer.
Other risks include false positives,
so finding a lung nodulethat's initially worrisome,
that might prompt additionaltests or even biopsies,
sometimes even surgery to remove it,
(10:39):
which end up being benign or negative,
and so that's a lot thatsomeone might go through.
There's a lot of anxietywith the tests as well.
And then, there's also a riskthat we would find things
that are abnormal thatwouldn't hurt somebody
that we end up treating.
And so because of these risks,
we don't recommend lung cancerscreening for everybody,
(11:03):
but there is some thoughtthat people outside
of the current criteriacould probably benefit,
and we need to do some more research
to figure out which groups those are.
- It sounds like there'sa potential to expand
but not expand to all,but more data is required.
- That's right.
If we screened everybodystarting at a young age,
(11:24):
we would find all lung cancers early
if that was somehow possible.
But besides our healthcare system
not being able to handle thatmuch screening and testing,
I think it probablywouldn't be a good idea
for people who are really at little risk.
It's hard because we do have outliers.
We see people who are very young
who wouldn't meet screening criteria,
(11:46):
and we've seen them diagnosedwith advanced lung cancer,
and it's hard not to wonder who else,
how could they have been screened?
Who else should be screened
to prevent these cases from happening?
- Is there a higherrisk around lung cancer
when you talk about Blackpopulations, Latino, Asians
or is that an independent variable?
- One of the biomedicalinformatics faculty here,
(12:09):
Summer Han, as well as one of our former
Stanford thoracic surgery fellows
who's now at MassachusettsGeneral Hospital, Jeff Yang,
both of them have done researchon the screening criteria
and how that affects disparitiesbetween African-Americans
and other racial populations in the US.
And so interestingly,
(12:30):
they found that using thenumber of years smoked
instead of pack-yearswould decrease disparities.
So they look at thingslike what percentage
of people in differentgroups with lung cancer
would have qualified for screening?
And so when they use the current criteria,
(12:50):
the proportion of AfricanAmericans who would qualify
is much lower than the proportionof Caucasian Americans.
So when they trieddifferent screening criteria
and used, for example,the number of years smoked
instead of pack-yearsand found that actually
a lot more of those AfricanAmerican people with lung cancer
would have qualified in that way.
(13:12):
As you can imagine, thisresearch is pretty difficult,
because there's one a set of research
where it looks at people withcancer, and so you can see
whether they would've met thescreening criteria or not.
But what even better is if youhave the entire denominator,
the cohort of healthy people
and the people who willdevelop lung cancer
(13:32):
and then see how many peoplewould you have to screen
in order to find a certain number
of people with lung cancer.
And so, Dr. Yang had looked at databases
in the southern United States.
Dr. Han has looked at databasesin California and Hawaii,
of course, very differentracial and ethnic makeups there.
But they have found similar things
(13:52):
that changing the criteria from pack-years
to years of smoking mightactually decrease the disparities
a lot among different groups.- That's incredible.
So, what do we do with this information?
What's the next step?
Because it sounds like theguidelines need to change.
- I know, I think it's really hard,
because it is hard to get this data,
(14:14):
and none of this dataaddresses the mortality benefit
in the (indistinct) atthis randomized trial
of 50,000 people did.
And that's not feasible to repeat that
for every group thatyou are interested in.
And so I'm very gladthat medical informatic
and statistician like Dr. Hanare looking at these databases
(14:36):
to try to extract the information
without doing a huge trial.
And I do think that,
as we continue withthese research avenues,
that we may be able to change guidelines.
It's slow but necessary work.
Another group that's affected
are Asians and Asian Americans.
We talked about never smokerswho develop lung cancer.
(14:57):
In the US, about 20% of peoplediagnosed with lung cancer
had never smoked.
And in Asia and amonggroups of Asian Americans,
it ranges all the way up to80%, it's very different.
It seems like a verydifferent type of disease.
And so because of that, thereare many countries in Asia
(15:19):
who are offering screening already
even to people who have never smoked.
So these guidelines are very different
than what we find in the US.
And one of the most developedstudies was in Taiwan,
called the TALENT study,and they enrolled people,
men and women who had never smoked,
who had one other risk factor,
so that could be secondhandsmoke or family history,
(15:42):
and they actually found thatthey detected lung cancer
in 2.1% of that group,which is even higher
than what was found in the heavy smokers
in the national lung screening trial.
And again, about half ofthe people had stage one,
so this was definitely findingmore early-stage disease.
(16:02):
And based on that study,
they started the Early Detection Program,
so they're currently screeningpeople with family history,
so never smokers with family history,
because they found that wasreally the biggest risk factor
for the people who had never smoked.
- That's a really greatsegue to my next question.
You have a really great example of Taiwan
(16:23):
and what we can learnfrom these approaches.
I completely understand thatthe United States is unique.
We are made up of a diverse population,
and so it's not a one-size-fits-all,
which it's part of its beauty,but it's also challenging
to address some of these health issues.
Are there any other countries
that have taken that similar approach
(16:45):
that have been successful or not?
- There are several countries in Asia
who have been screening never smokers,
but I think the largest one is in Taiwan,
easiest for us to lookat their publication.
It's been fascinatinglearning about screening
around the world.
There's a world conferenceon lung cancer every year
that the International Association
for the Study of Lung Cancer puts on,
(17:07):
and so they always havepeople from around the world
giving talks about screeningprograms in their countries
and what they're doing to implement them.
And it's really a very wide range
from countries who don't haveany screening programs at all
to ones that are just completely covered.
And so, I think the US, you know,
we know that we have a long way to go,
(17:27):
and we have ways tolearn from other places
in terms of implementation
and getting people who areeligible in for screening.
- We talked about low-dosecomputed tomography, LDCT.
Can you just give us alittle bit more insight?
How does it work and what has made it,
it sounds like the preferred method,
for lung cancer screening?
(17:49):
- Computing tomographyor CT scans, in general,
are a mix of X-rays andthen computer assistance.
And so essentially, it putstogether a very detailed picture
of the lungs and really the entire chest,
where we can see really tiny lung nodules
using these CT scans.
(18:10):
And the technology has gotten really good
so that the radiation exposure for LDCT
has become quite low, low enough
that we feel comfortablerecommending an annual scan
for people for screening.
I will add the low-doseCTs are really quick,
so they just take a few minutes.
They don't hurt at all.
(18:30):
People just have to climb onto the bed
and then they get moved into the scanner,
which is like a donut.
And for the LDCTs, there'sno needle involved,
no IV, no contrast, soit really is quite quick
and convenient for people.
- What were we doing before LDCTs?
- We were not doing any screening.
- Oh.- Yeah,
(18:51):
there were a lot of peoplelooking into chest X-rays,
which is one single film,rather than the LDCT,
which is many films in a lot more detail.
Some people looked into sputum samples,
so looking to see ifthere were tumor cells
in sputum that people would cough up.
But it wasn't until theNational Lung Screening Trial,
(19:12):
which was performed in theUS, then published in 2011,
that showed that therewas a mortality benefit.
So that trial, the NLST was sponsored
by the National Cancer Institute,
and it was a multicentertrial across the US
that enrolled over 50,000 people
and randomized them to eitherthe LDCT or the chest X-ray.
(19:37):
These were all heavy smokers,and they found that the group
that got the annual LDCT
had a 20% lower deathrate from cancer, 20%,
and so that's reallywhat sparked the change
in the guidelines in the US.
So that paper was published
in "The New England Journalof Medicine" in 2011.
(19:57):
The USPSTF recommended screening
based on those criteriaof the trial in 2013,
and then Medicare startedcovering it in 2015.
And since then, we'vemade some smaller changes
in the eligibility criteria.
There were some studies inEurope that showed that people
(20:19):
with slightly lower smokinghistory might benefit.
And so the USPSTF has revisedtheir criteria a few years ago
to take into account that new data.
- Well that's interesting,2015 is not long ago,
right, it's pretty new?
I am amazed by how innovative
(20:39):
and advanced medicine and clinical care is
and how it can be in so many areas,
but also, how it is not in so many areas.
- It's really fascinating tosee this change over time,
because I was essentially in training
and a new surgeon whenall of this was happening.
(21:01):
And so to see it in real timeand see how health policy
then reflects the medical research
has been really fascinating.
And a lot of people ask why surgeons
are involved with screening,
because it's really seenas a primary care arena,
and I think surgeons really saw the impact
that it made on patients,because I can operate
(21:22):
on someone with stage-one disease
and potentially offer them a cure.
By the time it gets tostage four and it's dense,
it's very rare that it can offer anything.
And so we love seeing screening,
and we're huge advocates for it,
because we want morepeople to be diagnosed
in that early stagewhen we can be helpful.
- What an incredible outlook too.
(21:44):
You've talked about the recent changes
in the screening guidelines.
I wanna pivot to some of thebarriers that patients face
when accessing lung cancer screening.
What have you seen, especiallyin underserved populations?
- That is a really great question,
because despite lung cancer being covered
(22:06):
for almost a decade now, ourscreening rates are really low.
In California, it'sestimated that only 16%
of eligible people are getting screened.
That's a lot of people that might benefit
that are not getting their annual screens.
We've looked into many of these barriers,
and I think that it isreally multifactorial.
(22:28):
I do feel like a lot of peoplestill don't know about it.
It's been almost a decade,but it's still really new.
We've done some surveysat community health fairs
and found that the percentage of people
that know about lung cancer screening
is far below people who know
about mammograms for breast cancer
or colonoscopies for colon cancer.
(22:49):
It is still just not widelyknown in the community.
I think another part isthat primary care physicians
are usually this firstaccess point to screening,
and they have been really overwhelmed.
It's hard to fit screening,
which is, by definition, noturgent, into a short visit
when you have many othermedical problems to discuss,
(23:11):
and so that's a big part of it.
And I think making it muchmore part of the workflow
for primary care physicians
and assisting them andsupporting them in that
is really important.
And then there are different factors
in terms of the health's policy,
which may be larger andwe could really change,
but it's pretty confusing toknow how much people will pay.
(23:32):
It all depends on their insurance.
Things are covered,but there are questions
about copay depending on insurance,
and I think a lot of that
just reflects our healthcare system,
and it's hard for us to answersome of those questions.
- I think you're spot on there.
I can see how all of thosefactors would impact.
I'm curious if you've also seen
(23:52):
whether socioeconomic factors,whether it has an impact
to adhering to lung cancerscreening guidelines.
Because I think what I heard from you
is that these are annual, right,
so it's not just a one and done?
- I don't know specificallyabout adherence,
but there have been a lot of studies
about disparities among groups,
(24:13):
racial and ethnic groups interms of lung cancer screening.
When you look back at theNational Lung Screening Trial,
which prompted all of this,
over 90% of the participants were White.
So about, I think less than5% were African American,
about 2% were Asian, soit doesn't totally reflect
(24:33):
the US makeup and even worldwide,then it really does not.
And then the studies that followed
were mostly done in Europe,
so they're really very differentin terms of racial makeup.
And one of the biggest onescalled the Nelson Trial,
they actually had about 85% men,
(24:54):
so women were a sub-analysis in the group.
So, these trials arehuge, and they gave us
a lot of information aboutthe mortality benefit,
but they definitely didn't answer
all of the questions that we had
in terms of who wouldbenefit from screening.
I think that it isreally hard to make sure
(25:14):
that people get screeningand know about it,
and then also, to adhereto the annual scan.
A lot of it is very easy,like the scan is very quick,
but it does requiretransportation to go get it,
if there are language barriers
in terms of understandingthe eligibility or adherence,
the time out of the day to go there,
(25:36):
so there are a lot of nuances that I think
would make adherence andfollow up a lot better.
- I think what you saidjust reflects the challenges
that exist in clinicaltrials and in general,
and not just for lung cancer,the need for more diversity
and more representationin clinical trials.
(25:58):
I think there's a growing recognition
of the need for that inthe scientific community.
I wonder whether psychological barriers
can also impact the patient
in terms of wanting togo through screening
where it's almost this notion
of ignorance is bliss type thing.
(26:19):
What are your thoughts on that?
- I think you are totally right.
I think it is really scary tothink about having lung cancer
and having a test that might find that
and really just change your entire life.
I think it's also hard,
because screening isbased on smoking history,
and we've known thatsmoking as a risk factor
(26:40):
for lung cancer for a decade now.
And so I think there is some stigma
in terms of being a smoker,
smoking despite knowingthat there is that risk,
and then potentiallyfeeling like you caused this
if you were smoking and have lung cancer.
Now a lot of people whosmoke don't get lung cancer,
(27:02):
and there are a lot ofpeople who have never smoked
who develop lung cancer.
So I try to reduce thatand really get people
to feel more open to screening,
but I can see that it is hard.
It's a very scary prospect.
- You alluded to this earlier
when we were talking about LDCT,
the false positive that can come up,
(27:23):
and I think it's one of the things
that it has been criticized for,
it's high false positive rates,potential over diagnosis.
How do clinicians balance this concern
when there are clearbenefits for early detection?
- Yeah, different clinicians
will respond differently to this question.
You're talking to one of the leads
(27:44):
of the lung cancer screeningprogram at Stanford,
but if you look at theNational Lung Screening Trial,
they quote a really high falsepositive rate of almost 25%,
which seems pretty ridiculous.
When you look at the details though,
they count even a nodulethat's four millimeters
as a positive, and so theseare nodules that are so small
(28:07):
that we wouldn't recommenddoing anything else
based on that nodule alone.
We would probably recommendanother scan in a year,
which is what is recommended anyway.
So it would still count as a positive
under the NLST definition,but wouldn't actually prompt
any invasive evaluation at that point.
(28:27):
So if you look at the falsepositive rate in NLST,
for the first couple years,it was around just under 25%,
and then for the third year,
suddenly that falsepositive rate drops to 7%,
and that's because if a nodulewas followed and didn't grow,
then it would be called positive initially
(28:48):
and then switched to negative.
And so it's interesting, Ithink that some of the numbers
of a high false positive rate
don't really reflect the risk to patients,
which is really the risk
when we decide to do more evaluation,
so invasive biopsies or testing
that might actually havecomplications that affect it.
(29:08):
- I think what I'm hearing you say,
and correct me if I'm wrong,
what I'm hearing you say is,"Do it anyway," and basically,
"Do a repeat LDCT before maybe taking
"any invasive measures."
- It depends on what is found.
American College ofRadiology actually came up
with a whole classificationsystem of lung nodules,
(29:31):
just to make sure thatthings were standardized,
and we could talk aboutthis false positive
or false negative rate.
They call it Lung-RADS,
lung and then reportingand data system, Lung-RADS,
and it goes from essentially zero to four,
and so each nodule that's found
is classified as one of these Lung-RADS.
(29:52):
And for each Lung-RADS,
they have an expectedpercentage that are malignant.
They expected prevalencewithin a population
and then the recommendations.
For example, if you have a Lung-RADS 2,
which is probably benign,
they would say get the scan in a year,
which is what someone would doif they didn't find a nodule.
(30:14):
And so, a lot of those small nodules
on NLST that were positiveactually would just go along
with this Lung-RADS classification,
and nothing would actuallybe different for that person.
Now as you get higher in Lung-RADS,
you get to ones that are very suspicious,
and over 50% of those willbe diagnosed as lung cancers.
(30:37):
And those are the onesthat you'd recommend,
they call it tissue sampling or biopsy,
and that can even involve surgery.
And so I think that is when
we have a much more detaileddiscussion with a patient
about the options and the risks
for these invasive procedures.
- That's very helpful to clarify.
I'm going to revise what I'mhearing from you at this point.
(30:59):
Do it anyway, but your analysis
of this quote-unquote false positive
is a little bit more nuanced,
and there's additional classifications
that you need to utilize todetermine what to do with that.
- The vast majority of peoplewho have lung nodules found
(31:19):
many times what we dofor those lung nodules,
especially when they'rereally small, is just watch.
When we find really small lung nodules,
they're too small tobiopsy or test in any way,
and so the recommendationis just to follow them.
So in that sense, if thatcounts as a false positive,
(31:39):
that's something that won'tnecessarily harm someone.
I think where we're worriedabout finding a false positive
is when they undergo procedures,
like a bronchoscopy, abiopsy, again, even surgery.
Sometimes I will remove a nodule
that looks really suspicious with surgery,
because I'm so worried about it.
(32:01):
And we celebrate when it'snothing, but in that sense,
that patient didn't need it,we just didn't know for sure.
It is fair that we don'twant to expand screening
to everybody, 20-year-old, veryunlikely to have lung cancer
and any nodule found is muchmore likely to be benign.
So in that sense, screening someone
(32:22):
who's clearly at extremely lowrisk doesn't make any sense,
because there'd be a lot of anxiety,
potential other procedures or tests
when most likely, not lung cancer.
The question is where to draw that line,
and I think that atleast for heavy smokers,
it's been answered.
The NLST, despite thesefalse positive rates,
(32:44):
showed a mortality benefit,
so that is an easy,"Yes, go get screened."
And the other question is other groups,
where we think they're at higher risk,
but it's a little bit more unclear
what that mortality benefit would be.
That's where it's a little bit harder,
and we need more research.
- Are there any emerging screening methods
or technologies that could either enhance
(33:07):
or complement some of the LDCTs
in our current lung cancerscreening practices?
- I think there are a couple avenues.
One is that the technology for CT scans
keeps getting better, andso some of our radiologists,
Henry Guo is one of them here at Stanford,
is looking into photon counting CT scans
(33:29):
and whether they couldadminister screening tests
with even lower dose CT,
so we called that ultra-low dose CT scan.
And then blood-based tests area huge upcoming technology.
We had one of our radiationoncologist, Max Diehn here,
who's been doing research intesting, circulating tumor DNA
(33:52):
in blood sample for many years.
And I think those will beamazing when we could do,
that'll be even betterthan getting a CT scan
if we could do that for lung cancer
and potentially other cancers.
- That really sounds exciting.
In the past few years at least,
AI is everything everyone's talking about.
Has that been in the lungcancer screening space as well?
(34:15):
Have we been talkingabout applications for AI
in lung cancer screening?- Yes, for sure.
There have been a couple different ways
people have started looking at AI,
but I think that there is alot more to go in the future.
Reading the CT scan, Ithink, is one big part,
so finding lung nodules, we'refollowing them over time.
It's something that takes a lot of skill
(34:37):
from our radiologists andso trying to figure out
if AI can be helpfulin that would be great.
And then I think also looking at risk
for lung cancer in general
would be another potential application.
None of these are beingused right now consistently,
so I think, again, we have alot more to learn about them,
but I'm very excited aboutpotential applications.
(35:00):
- Looking forward, what do you envision
as the future of lung cancer screening?
- That is a really good question.
I am hoping that one,
we get people who weknow are eligible now,
we know there's a mortality benefit,
that we find a way to reachthem and get them screened.
I think that's the most important thing.
(35:20):
And then second, we know thatother people will benefit,
and we're just not sure who exactly,
and so I think that researchwill be really important
to figure out how else we cansave lives with screening.
And then the last thing,which you already asked about
is other methods of screening.
I think in the future,these blood-based tests,
(35:43):
any tests that will make it easier
for people to get screened and diagnosed
will be better and really help save lives.
- This has been a wonderful conversation
with so many key takeaways for clinicians,
whether you're a primary care clinician
or a lung cancer specialist.
I so appreciate your time today.
- Thank you so much, Dr.Adewuya. It's been great.
(36:03):
(gentle music)
- This episode was broughtto you by Stanford CME.
To claim CME forlistening to this episode,
click on the Claim CME link below
or visit Medcast.stanford.edu.
Check back for new episodes
by subscribing to Stanford Medcast
wherever you listen to podcasts.
(36:24):
(bright music)
Popular Podcasts
Stuff You Should Know
If you've ever wanted to know about champagne, satanism, the Stonewall Uprising, chaos theory, LSD, El Nino, true crime and Rosa Parks, then look no further. Josh and Chuck have you covered.
Dateline NBC
Current and classic episodes, featuring compelling true-crime mysteries, powerful documentaries and in-depth investigations. Follow now to get the latest episodes of Dateline NBC completely free, or subscribe to Dateline Premium for ad-free listening and exclusive bonus content: DatelinePremium.com
CrimeLess: Hillbilly Heist
It’s 1996 in rural North Carolina, and an oddball crew makes history when they pull off America’s third largest cash heist. But it’s all downhill from there. Join host Johnny Knoxville as he unspools a wild and woolly tale about a group of regular ‘ol folks who risked it all for a chance at a better life. CrimeLess: Hillbilly Heist answers the question: what would you do with 17.3 million dollars? The answer includes diamond rings, mansions, velvet Elvis paintings, plus a run for the border, murder-for-hire-plots, and FBI busts.