April 18, 2025 • 22 mins
In this episode of Lab Medicine Rounds, Justin Kreuter, M.D., and Miglena Komforti, D.O., from Mayo Clinic as they discuss threshold diagnoses of the breast — those gray zones that challenge even seasoned professionals. Learn why they’re important, how to approach them, and what role technology might play in the future.
Mark as Played
Transcript
Episode Transcript
Available transcripts are automatically generated. Complete accuracy is not guaranteed.
Justin Kreuter, M.D. (00:07):
This is Lab Medicine Rounds, a curated
podcast for physicians,laboratory professionals, and
students. I'm your host, JustinKreuter, a transfusion medicine
pathologist and assistantprofessor of laboratory medicine
and pathology at Mayo Clinic.Today, we're rounding with
assistant professor oflaboratory medicine pathology
from Mayo Clinic Florida'scampus, doctor Komforti, to talk
(00:30):
about navigating thresholddiagnosis in the breast. Thanks
for joining us today, doctorKomforti.
Miglena Komforti, D.O. (00:37):
Thank you for having me. It's
wonderful to be here.
Justin Kreuter, M.D. (00:40):
So I really appreciate being able to
dive into this anatomicpathology topic as I'm still
anatomic board certified, soyou're gonna help keep me fresh
today. But maybe for a morebroad audience, why is this
topic of navigating thresholddiagnoses? Why is it an
important one to talk about?
Miglena Komforti, D.O. (01:01):
Sure. Sure. Great question. And really
taking a step back and lookingat the big picture, anatomic
pathology, surgical pathologyremains the gold standard of
tissue diagnosis, and uponwhich, really, we build
management for our patients.Right?
So in very layman terms, if youthink of a patient who presents
(01:21):
with a breast mass or onradiology on screening
mammogram, there's a mass, wedon't know benign, malignant, or
anything more than just there'sa mass. So we have to get a
biopsy. We have to get tissue,and it's on the pathologist to
make that precise, accuratediagnosis to guide the clinical
team.
Justin Kreuter, M.D. (01:41):
Wow. So that really puts a nice fine
point on today's conversation,right? I mean, you're talking
about something that is benign,something that is malignant. I
mean, these are two verydifferent forks in the road and
highlighting that's your role asan anatomic pathologist. What
makes a threshold diagnosis?
(02:02):
And I assume when we saythreshold diagnosis, we're
saying something that goes fromsomething that might be more
benign to a malignancy. Whatmakes that a challenge?
Miglena Komforti, D.O. (02:13):
Yeah, absolutely. And I really would
like to just, again, take a stepback and sort of put things into
perspective for the wideaudience range that we have.
It's important to know what evena threshold diagnosis is. It's
somewhat of a newer term. Youmay hear in the literature
borderline lesion in the breast,that's another way people have
(02:33):
referred to this kind ofdiagnosis.
So what are we really talkingabout here? So back to
pathologists rendering aninterpretation. It is an
interpretation. So it is throughthe lens, through the prism of
that individual. That's oneelement.
The second element is that thereare certain diagnoses in the
breast that just have inherentdiagnostic variability. And part
(02:58):
of that is the fact that we havean imperfect tool, and that is
the eye. Right? So the eye canonly see that far, we cannot see
molecular alterations with justthe naked eye. So some of those
entities have overlappingmorphology, they have
overlapping immunophenotypicalfindings, they have overlapping
(03:20):
molecular findings, and again,back to the eye, not being able
to quite discern between oneentity and the other.
And in general, when we lookback over the past fifty years
of breast care, we've reallychanged the way we manage our
patients. In the 1990s, we weredoing mastectomies for anything
and everything, and now we'veshifted to breast conservation
(03:43):
surgeries, to sentinel lymphnode biopsies rather than
axillary lymph node dissection.Even trials at this time that
are surveillance only trials forpeople with ductal carcinoma in
situ of the breast. That meanswe're leaving this non obligate
precursor lesion inside thewoman or the man's breast and
(04:06):
we're observing only. So thingshave shifted quite a lot, but
it's really dependent back on usas pathologists to make that
diagnosis to help our clinicalcolleagues to know which route
to go.
So I mentioned a couple ofentities here, and I'll just go
back and review those. One ofthem was ductal carcinoma in
(04:27):
situ, low grade ductal carcinomain situ. That is a threshold
lesion. It has overlappingfeatures with atypical ductal
hyperplasia. Some of us in thepathology world, my colleagues
know, we refer to those as ADHand DCS.
And our criteria for those twolesions, how to discern them
(04:47):
adequately and accurately, thatcriteria was developed in 1982
and later on with an update in1990. And now it's 2025. So what
we used to do before fifty yearsago, forty years ago, doesn't
quite give us enough informationin the era that we are
(05:08):
currently, which is the era ofde escalation of treatment for
breast cancer patients. So howdoes that again relate to
threshold lesions? Why is thatimportant?
Because we are increasinglyasked to give more and more
details to continue substratifying into more and more
diagnostic buckets all theselesions and providing more
(05:28):
detail to our clinicians so theycan pick the correct route of
treatment. And I'll just stopthere and give you an
opportunity because I realize Icovered a lot of different
topics in a very short amount oftime. So I'm ready for a
question if you have one.
Justin Kreuter, M.D. (05:43):
Yeah, yeah, I really appreciate how
you took us through there andwe're talking about the example
of ductal carcinoma in situ andthe very closely similar but
different diagnosis of atypicalductal hyperplasia. The DCIS,
ADH sort of dilemma. If I'mthinking about our our student
(06:06):
listeners out there, kind ofappreciating the fact that this
kind of highlights the reasonwhy a pathologist is somebody
that has to be a physician tounderstand what is the clinical
significance of what aparticular call is, well as your
specialized skills as apathologist. Is that something
that when you were training, youwere really cognizant of
(06:28):
connecting your medicalbackground and your pathology
specific training?
Miglena Komforti, D.O. (06:34):
Right. Great question. Extremely
important to know what yourdiagnosis means to the patient,
to the clinician. I would sayfor our students, our residents
just starting in pathology, getthe basics down first. Get to
understand those ideal images ofwhat ADH looks like, of what DCS
(06:57):
looks like, what invasivecarcinoma.
Get to know that, build thatbrain bank of images, and then
be prepared that depending onthe practice you go to, where
you are, you may have asubstantial number of those
threshold lesions, perhaps evenup to fifteen percent. So a
fraction of what you're going tosee will fall into that, oh my
(07:18):
god, I'm not quite sure. Is itADH or is it DCIS? So I think,
of course, know the goodexamples because we always
compare abnormal to normal. Soyou have to know in your mind
what abnormal looks like to makethat determination.
And that's how I approach mycareer. That's how I approach my
(07:40):
training. And then it was thatextra layer of learning, and I
would say that learning neverreally stops. It continues. It's
a lifelong learning journey ofwhat it means to make the
diagnosis and how to fine tuneand how to find that sweet spot
of feeling comfortable withuncertainty.
And that's what really thresholdlesions are. They're
(08:02):
uncertainty, and you, as abreast pathologist, have to find
that comfort level.
Justin Kreuter, M.D. (08:07):
I really appreciate you use that phrase.
It's kind of navigatinguncertainty. That's something
that I've I've hear discussed inmany contexts of medical care.
If we kind of putting onimagining now, not necessarily a
student, but somebody who'smaybe young in their career in
this podcast, what are yourthoughts on recommendations for
(08:29):
how do we navigate thesethreshold diagnosis? You're
confronted with a case that's,you know, is this DCIS?
Is this ADH as you've beentalking about? How might one
start to navigate that well?
Miglena Komforti, D.O. (08:42):
Right. Great question. I think, again,
back to what we just said iscover the basics. Right? Cover
the basics.
Because if you don't cover themand you show this to your
colleagues, they will ask you dothe basics. What are the basics?
The basics are let's stain it.That is an easy thing to do.
Common stains that almost everylab has.
We certainly have them here inMayo and all of our campuses.
(09:05):
Get that CK five, keratin five,get that estrogen receptor
stain, and see what the lesionis staining like. Of course,
there are other things you cantry, levels. We like to think of
lesions as two d, but inreality, they're three d. It's a
formalin fixed paraffin embeddedblock that is approximately four
(09:26):
or five millimeters, so there'sdefinitely that Z plane, right?
Let's see how the tissue, howthe lesion develops. Is it going
to open up? Is it going to getsmaller? Did we pretty much
excise it with the biopsy orthere's more that we're not
seeing? So get those levels, getthat basic workup.
That's the number one. Peoplesay, my more senior colleagues
will say, I want to sleep on it.Right? So just think of the next
(09:50):
day, look at those stains, andcome back with fresh eyes. And I
like to also say it's mostimportant to get the diagnosis
accurate rather than to rush it.
So let your clinical colleaguesknow that there may be a little
bit of delay on the case becausewe're working it up. That is
perfectly appropriate. So that'sthe basics. Now what else can
(10:10):
you do to help yourself? Andthere are a few things you can
do to help yourself.
Number one, don't be a hero.Don't commit to a diagnosis that
is not quite applicable to thispatient. So if you need to have
a descriptive diagnosis on yourfinal diagnosis, then utilize
(10:30):
that. What are some examples? Wekeep talking about ADH and low
grade TCS.
We know there's going to be asubset of those that will define
accurate reproducibleclassification. So when you're
faced with one of those, becauseit's a matter of when, not if,
when you're faced with one ofthose, use a descriptive
diagnosis. For example, atypicalintraductal epithelial
(10:52):
proliferation, C comment, or atleast atypical ductal
hyperplasia, or atypical ductalhyperplasia bordering ductal
carcinoma in situ. Let yourcolleagues know that this is not
your straightforward lesion. Andthat is a way to help yourself
because we don't want toovercommit, and we don't want to
be too brave with our diagnosisbecause you can imagine the
(11:13):
significance of carryingdiagnosis of cancer that comes
with a lot of repercussions.
Right? A a lot of anxiety forthe patient. There's a lot of
treatment that's involved, andso on and so forth. So you're
helping yourself, you're helpingthe patient, you're helping your
clinical colleagues by admittingthe challenges of the lesion.
(11:35):
For more laboratory education,including a listing of
conferences, webinars, and ondemand content, visit
MayoClinicLabs.com/education.
Justin Kreuter, M.D. (11:49):
If I could interject for a
Miglena Komforti, D.O. (11:51):
quick
Justin Kreuter, M.D. (11:51):
second and just say, this conversation is
reminding me about why I wentinto pathology in the first
place, right? The it's an honestmedicine, right, where you're
saying, like, you're not beingthe hero of kinda flat footed
calling something, but you takethe histology to the limits.
What you were able to call, youcall, but you're careful to not
(12:12):
go further.
Miglena Komforti, D.O. (12:13):
That's right. That's why we have to we
have to practice safe medicinefirst and foremost. Right? And,
of course, it's up to thepatient what kind of treatment
they'll elect at the end, but Ihave to be able to sleep at
night to know that I've done thebest I can. And sometimes that
means, in this day and age, wementioned de escalation of
(12:34):
therapy.
We've removed ourselves frombeing too aggressive in surgery,
in oncology, in radiationtreatment. And so in pathology
as well, what we used to by thepage criteria that we mentioned
1982 or so, 1985, around thattime, that criteria tells us
that as long as we have twomillimeters to two duct
involvement by these low gradeneoplastic proliferation of
(12:57):
cells, then we feel comfortablecalling it DCS. In reality, I
will tell you, myself included,my senior colleagues in
pathology, we actually allow formore. We're pushing that border.
We're moving that bar.
We allow for three millimeters,maybe four millimeters. We like
to quote see more, and there'snot really a a good definition
(13:17):
of more look what more lookslike, but we are pushing that
bar. We're being safer. We arenoticing that with all the
advancement that's happened inthe past, again, forty, fifty
years in breast medicine, breastcancer mortality increases. So
we've been excellent in catchingsmaller and smaller lesions.
(13:39):
We're talking about couplemillimeters. Our radiology
colleagues are phenomenal. Theycatch small calcifications,
small lesions. We excise them.We treat the patient, and yet
there's no really effect onmortality from invasive breast
cancer.
And that's really what we'retrying to do. We're trying to
prevent breast cancer bytreating ADH, atypical ductal
hyperplasia, and low grade DCSductal carcinoma in situ, the
(14:03):
nonobligate precursor. Right? Somaybe we're overtreating or
maybe we're overdiagnosing theselesions and being a little too
aggressive. So back to don't bea hero.
You're doing service to yourselfand the patient by being careful
and admitting challenges. Andthat I would also go on a little
(14:23):
bit of a side note here andmention when you admit
challenges, when you have acommon that explains, hey, this
is an in between lesion, this isa borderline, a threshold
lesion, you prevent themisconception that another
pathologist is wrong. Becauseoftentimes what patients do, and
they're extremely smart, theywill go to another institution,
(14:46):
they will seek a second opinion.And depending on how the other
pathologist is trained,remember, going back to this is
just an interpretation, dependson the person interpreting it.
Right?
The pathologist. Depending howthat person is trained in their
personal approach andunderstanding of the literature,
they may be just a little moreaggressive. And then two
different diagnosis may happen.And then the patients may think,
(15:08):
well, you know, one pathologistis wrong. Which one is it?
These are two separate lesions.And there's a lot more
understanding than that that wedon't convey, that these are
related lesions from the lowgrade neoplasia pathway and that
they're threshold lesions.Right? So by being careful and
not being a hero, you're helpingyour colleagues and prevent the
misconception of somebody, apathologist, being wrong.
Justin Kreuter, M.D. (15:31):
I really appreciate that. And I think it
underscores for the clinicianlisteners' podcast, right? If
you see a pathologist that'susing a descriptive diagnosis,
maybe they're using a diagnosisthat you're unfamiliar with or
like doctor Komforti mentioned,basically using see comment in
the diagnosis field. I thinkit's highlighting that
(15:53):
definitely see the comment andalso don't hesitate to reach out
with a phone call.
Miglena Komforti, D.O. (15:59):
Yeah. That's right. When we write
these reports, we write them fora wide audience. So, obviously,
the patient, obviously, thesurgeon, radiologist, but
there's billing. There'sresearch.
There's the other pathologistwho, again, if the case goes
out, the material is sent toanother institution. There's the
trainee. I talk to myselfsometimes. I use certain
(16:21):
terminology and certainsentences. So in a situation
where I get a phone call,anybody asking about my report,
I can tell you exactly what Isaw because I understand what my
words mean to me.
We speak to a wide range ofpeople, the nurses or the techs
who code these lesions. Right?They have to have a drop down
menu. They have to codeeverything. It's very
(16:42):
complicated.
So every so often, things willbe conveyed in a way that's
perhaps suboptimal because it'snot a diagnostic bucket, and we
like buckets because they followan algorithm, a treatment
algorithm. But we also have toadmit that breast pathology has
those threshold diagnosis andlesions, and it's just by
(17:06):
playing the numbers. Eventually,you'll be faced with one of
them.
Justin Kreuter, M.D. (17:09):
And I appreciate you've taken us
through a little bit of, youknow, how to work through these
challenges. Kind of putting onmy looking into the future hat,
what do you think? I mean, isthis something that, you know, I
think you mentioned, right, it'sbeen several decades since the
criteria was previously revised.I mean, is this example that
you've mentioned today, thisductal carcinoma in situ versus
(17:32):
atypical ductal hyperplasia, isthis threshold diagnosis just
going to continue to be adiagnostic challenge or there's
gonna be new technologies or newways that we may separate these
more reliably, and maybethere'll be different challenges
going forward in the future.What do you see on your future
(17:55):
prognosticating for how how thisis gonna look?
Miglena Komforti, D.O. (17:58):
I think it's very exciting. You know,
some of my colleagues who arelistening or for the students
and residents who are thinkingbreast pathology, they may be
discouraged because they'rethinking, oh my god. This is so
hard. And it is hard, but it'sexciting because there's so much
more work to be done, so muchmore to be discovered. It's
truly, we don't know, right?
And you could be the next personwho cures cancer, I don't know.
(18:22):
But I do think there is going tobe a shift, and I think part of
that is AI. We've seen AI helpus on many different things and
fronts, in breast pathologyincluded in Mayo Clinic Florida.
We do automated interpretationwith AI for some of the
biomarkers in breast cancer, forexample, proliferation index,
MIB. In Rochester, they doestrogen receptor, progesterone
(18:44):
receptor HER2.
So it's extremely helpfulbecause it provides that
precision. It is not asubjective, it is an objective
tool. It's reproducible, and itprovides precision. I think
there's going to be animprovement in the way we render
diagnosis just by incorporatingsome of those AI tools. But I
(19:05):
also think that there has to bean added layer, perhaps
incorporating molecular into ourdiagnosis and not just
immunophenotypic findings andmorphology.
I think there has to be an addedlayer, and I know there are a
couple of companies out thereworking on that. And hopefully
in a couple of years, if we dothis again, I may have some more
(19:26):
information for you that, youknow, we can share. But as of
right now, work in progress.
Justin Kreuter, M.D. (19:32):
Right on. We'll definitely be bringing you
back. One final question I havefor you since you brought up and
we're talking about AI. As apathologist, I work a lot with
the medical school. Mhmm.
And I think that pathology isone of these careers that I
think there's a perception amongmedical students of, in the
world of AI, you know, what is apathologist going to do? What
(19:55):
are your thoughts as an atomicpathologist in AI? You know,
your answer there, youhighlighted, was really talking
about it kind of almost behavingas an adjunct of kind of
integrating additionalinformation, then you as the
pathologist overseeing that. Butwhat do you think that future
interface is going to look like?
Miglena Komforti, D.O. (20:16):
Right. This is a tale as old as time.
So it I've been told by mysenior colleagues that when
immunohistochemistry came out,people were pathologists were
like, oh my god. This isthreatening our jobs. And then
molecular came out, andpathologists were like, oh my
god.
It's threatening our jobs. So inother words, this will be the
(20:37):
solution to our problems, and itwas not. And I kind of hear a
little bit of that with AI whereAI is out and it's going to
solve all of our problems andpathologists will be out of
jobs. Right? Something thatpeople forget is that it is the
breast pathologist, and becausewe're talking about breast
pathology, but it's the breastpathologist who teaches that AI.
(21:00):
So who builds the ground truthfor the AI? It is that
pathologist. And I've had acouple of projects where we sit
down and we annotate and weteach the algorithm what cancer
looks like, what benign lookslike. So there has to be a human
element. There has to be a humanelement.
And I think we need a lot ofprospective data before we can
(21:22):
remove the human element. Untilthen, I think we'll be okay.
Justin Kreuter, M.D. (21:28):
You know, I I totally agree with you and I
see that that role and and Ithink my own thought on this is
you're highlighting this abilitythat it's going to enable you to
focus on even other thingsmaybe. And and as you pointed
out in in the specific case ofbreast cancer, maybe what you
are able to focus on goingforward is gonna be something
(21:48):
that moves the needle as far aswhat does the mortality look
like for men and women whosuffer from breast cancer.
Miglena Komforti, D.O. (21:55):
That's right. And there's a lot to do
in pathology, and I urge you toexplore it as students or or
residents interested in thatspecialty. It's not just sitting
in your office looking at slidesall day. It's very exciting.
There's a lot of, like wementioned, AI.
There's education. There'sresearch, trials, innovation. It
is just a very exciting time.
Justin Kreuter, M.D. (22:17):
We've been rounding with Dr. Komforit
talking about navigatingthreshold diagnoses of the
breast. I really appreciate youjoining us today.
Miglena Komforti, D.O. (22:25):
Thank you. My pleasure.
Justin Kreuter, M.D. (22:26):
And to our listeners, thank you for joining
us today. We invite you to shareyour thoughts and suggestions
via email tomcleducation@mayo.edu. If you've
enjoyed this podcast, pleasesubscribe. And until our next
rounds together, we encourageyou to continue to connect lab
medicine and the clinicalpractice through educational
(22:47):
conversations.
This is Lab Medicine Rounds, a curated
podcast for physicians,laboratory professionals, and
students. I'm your host, JustinKreuter, a transfusion medicine
pathologist and assistantprofessor of laboratory medicine
and pathology at Mayo Clinic.Today, we're rounding with
assistant professor oflaboratory medicine pathology
from Mayo Clinic Florida'scampus, doctor Komforti, to talk
(00:30):
about navigating thresholddiagnosis in the breast. Thanks
for joining us today, doctorKomforti.
Miglena Komforti, D.O. (00:37):
Thank you for having me. It's
wonderful to be here.
Justin Kreuter, M.D. (00:40):
So I really appreciate being able to
dive into this anatomicpathology topic as I'm still
anatomic board certified, soyou're gonna help keep me fresh
today. But maybe for a morebroad audience, why is this
topic of navigating thresholddiagnoses? Why is it an
important one to talk about?
Miglena Komforti, D.O. (01:01):
Sure. Sure. Great question. And really
taking a step back and lookingat the big picture, anatomic
pathology, surgical pathologyremains the gold standard of
tissue diagnosis, and uponwhich, really, we build
management for our patients.Right?
So in very layman terms, if youthink of a patient who presents
(01:21):
with a breast mass or onradiology on screening
mammogram, there's a mass, wedon't know benign, malignant, or
anything more than just there'sa mass. So we have to get a
biopsy. We have to get tissue,and it's on the pathologist to
make that precise, accuratediagnosis to guide the clinical
team.
Justin Kreuter, M.D. (01:41):
Wow. So that really puts a nice fine
point on today's conversation,right? I mean, you're talking
about something that is benign,something that is malignant. I
mean, these are two verydifferent forks in the road and
highlighting that's your role asan anatomic pathologist. What
makes a threshold diagnosis?
(02:02):
And I assume when we saythreshold diagnosis, we're
saying something that goes fromsomething that might be more
benign to a malignancy. Whatmakes that a challenge?
Miglena Komforti, D.O. (02:13):
Yeah, absolutely. And I really would
like to just, again, take a stepback and sort of put things into
perspective for the wideaudience range that we have.
It's important to know what evena threshold diagnosis is. It's
somewhat of a newer term. Youmay hear in the literature
borderline lesion in the breast,that's another way people have
(02:33):
referred to this kind ofdiagnosis.
So what are we really talkingabout here? So back to
pathologists rendering aninterpretation. It is an
interpretation. So it is throughthe lens, through the prism of
that individual. That's oneelement.
The second element is that thereare certain diagnoses in the
breast that just have inherentdiagnostic variability. And part
(02:58):
of that is the fact that we havean imperfect tool, and that is
the eye. Right? So the eye canonly see that far, we cannot see
molecular alterations with justthe naked eye. So some of those
entities have overlappingmorphology, they have
overlapping immunophenotypicalfindings, they have overlapping
(03:20):
molecular findings, and again,back to the eye, not being able
to quite discern between oneentity and the other.
And in general, when we lookback over the past fifty years
of breast care, we've reallychanged the way we manage our
patients. In the 1990s, we weredoing mastectomies for anything
and everything, and now we'veshifted to breast conservation
(03:43):
surgeries, to sentinel lymphnode biopsies rather than
axillary lymph node dissection.Even trials at this time that
are surveillance only trials forpeople with ductal carcinoma in
situ of the breast. That meanswe're leaving this non obligate
precursor lesion inside thewoman or the man's breast and
(04:06):
we're observing only. So thingshave shifted quite a lot, but
it's really dependent back on usas pathologists to make that
diagnosis to help our clinicalcolleagues to know which route
to go.
So I mentioned a couple ofentities here, and I'll just go
back and review those. One ofthem was ductal carcinoma in
(04:27):
situ, low grade ductal carcinomain situ. That is a threshold
lesion. It has overlappingfeatures with atypical ductal
hyperplasia. Some of us in thepathology world, my colleagues
know, we refer to those as ADHand DCS.
And our criteria for those twolesions, how to discern them
(04:47):
adequately and accurately, thatcriteria was developed in 1982
and later on with an update in1990. And now it's 2025. So what
we used to do before fifty yearsago, forty years ago, doesn't
quite give us enough informationin the era that we are
(05:08):
currently, which is the era ofde escalation of treatment for
breast cancer patients. So howdoes that again relate to
threshold lesions? Why is thatimportant?
Because we are increasinglyasked to give more and more
details to continue substratifying into more and more
diagnostic buckets all theselesions and providing more
(05:28):
detail to our clinicians so theycan pick the correct route of
treatment. And I'll just stopthere and give you an
opportunity because I realize Icovered a lot of different
topics in a very short amount oftime. So I'm ready for a
question if you have one.
Justin Kreuter, M.D. (05:43):
Yeah, yeah, I really appreciate how
you took us through there andwe're talking about the example
of ductal carcinoma in situ andthe very closely similar but
different diagnosis of atypicalductal hyperplasia. The DCIS,
ADH sort of dilemma. If I'mthinking about our our student
(06:06):
listeners out there, kind ofappreciating the fact that this
kind of highlights the reasonwhy a pathologist is somebody
that has to be a physician tounderstand what is the clinical
significance of what aparticular call is, well as your
specialized skills as apathologist. Is that something
that when you were training, youwere really cognizant of
(06:28):
connecting your medicalbackground and your pathology
specific training?
Miglena Komforti, D.O. (06:34):
Right. Great question. Extremely
important to know what yourdiagnosis means to the patient,
to the clinician. I would sayfor our students, our residents
just starting in pathology, getthe basics down first. Get to
understand those ideal images ofwhat ADH looks like, of what DCS
(06:57):
looks like, what invasivecarcinoma.
Get to know that, build thatbrain bank of images, and then
be prepared that depending onthe practice you go to, where
you are, you may have asubstantial number of those
threshold lesions, perhaps evenup to fifteen percent. So a
fraction of what you're going tosee will fall into that, oh my
(07:18):
god, I'm not quite sure. Is itADH or is it DCIS? So I think,
of course, know the goodexamples because we always
compare abnormal to normal. Soyou have to know in your mind
what abnormal looks like to makethat determination.
And that's how I approach mycareer. That's how I approach my
(07:40):
training. And then it was thatextra layer of learning, and I
would say that learning neverreally stops. It continues. It's
a lifelong learning journey ofwhat it means to make the
diagnosis and how to fine tuneand how to find that sweet spot
of feeling comfortable withuncertainty.
And that's what really thresholdlesions are. They're
(08:02):
uncertainty, and you, as abreast pathologist, have to find
that comfort level.
Justin Kreuter, M.D. (08:07):
I really appreciate you use that phrase.
It's kind of navigatinguncertainty. That's something
that I've I've hear discussed inmany contexts of medical care.
If we kind of putting onimagining now, not necessarily a
student, but somebody who'smaybe young in their career in
this podcast, what are yourthoughts on recommendations for
(08:29):
how do we navigate thesethreshold diagnosis? You're
confronted with a case that's,you know, is this DCIS?
Is this ADH as you've beentalking about? How might one
start to navigate that well?
Miglena Komforti, D.O. (08:42):
Right. Great question. I think, again,
back to what we just said iscover the basics. Right? Cover
the basics.
Because if you don't cover themand you show this to your
colleagues, they will ask you dothe basics. What are the basics?
The basics are let's stain it.That is an easy thing to do.
Common stains that almost everylab has.
We certainly have them here inMayo and all of our campuses.
(09:05):
Get that CK five, keratin five,get that estrogen receptor
stain, and see what the lesionis staining like. Of course,
there are other things you cantry, levels. We like to think of
lesions as two d, but inreality, they're three d. It's a
formalin fixed paraffin embeddedblock that is approximately four
(09:26):
or five millimeters, so there'sdefinitely that Z plane, right?
Let's see how the tissue, howthe lesion develops. Is it going
to open up? Is it going to getsmaller? Did we pretty much
excise it with the biopsy orthere's more that we're not
seeing? So get those levels, getthat basic workup.
That's the number one. Peoplesay, my more senior colleagues
will say, I want to sleep on it.Right? So just think of the next
(09:50):
day, look at those stains, andcome back with fresh eyes. And I
like to also say it's mostimportant to get the diagnosis
accurate rather than to rush it.
So let your clinical colleaguesknow that there may be a little
bit of delay on the case becausewe're working it up. That is
perfectly appropriate. So that'sthe basics. Now what else can
(10:10):
you do to help yourself? Andthere are a few things you can
do to help yourself.
Number one, don't be a hero.Don't commit to a diagnosis that
is not quite applicable to thispatient. So if you need to have
a descriptive diagnosis on yourfinal diagnosis, then utilize
(10:30):
that. What are some examples? Wekeep talking about ADH and low
grade TCS.
We know there's going to be asubset of those that will define
accurate reproducibleclassification. So when you're
faced with one of those, becauseit's a matter of when, not if,
when you're faced with one ofthose, use a descriptive
diagnosis. For example, atypicalintraductal epithelial
(10:52):
proliferation, C comment, or atleast atypical ductal
hyperplasia, or atypical ductalhyperplasia bordering ductal
carcinoma in situ. Let yourcolleagues know that this is not
your straightforward lesion. Andthat is a way to help yourself
because we don't want toovercommit, and we don't want to
be too brave with our diagnosisbecause you can imagine the
(11:13):
significance of carryingdiagnosis of cancer that comes
with a lot of repercussions.
Right? A a lot of anxiety forthe patient. There's a lot of
treatment that's involved, andso on and so forth. So you're
helping yourself, you're helpingthe patient, you're helping your
clinical colleagues by admittingthe challenges of the lesion.
(11:35):
For more laboratory education,including a listing of
conferences, webinars, and ondemand content, visit
MayoClinicLabs.com/education.
Justin Kreuter, M.D. (11:49):
If I could interject for a
Miglena Komforti, D.O. (11:51):
quick
Justin Kreuter, M.D. (11:51):
second and just say, this conversation is
reminding me about why I wentinto pathology in the first
place, right? The it's an honestmedicine, right, where you're
saying, like, you're not beingthe hero of kinda flat footed
calling something, but you takethe histology to the limits.
What you were able to call, youcall, but you're careful to not
(12:12):
go further.
Miglena Komforti, D.O. (12:13):
That's right. That's why we have to we
have to practice safe medicinefirst and foremost. Right? And,
of course, it's up to thepatient what kind of treatment
they'll elect at the end, but Ihave to be able to sleep at
night to know that I've done thebest I can. And sometimes that
means, in this day and age, wementioned de escalation of
(12:34):
therapy.
We've removed ourselves frombeing too aggressive in surgery,
in oncology, in radiationtreatment. And so in pathology
as well, what we used to by thepage criteria that we mentioned
1982 or so, 1985, around thattime, that criteria tells us
that as long as we have twomillimeters to two duct
involvement by these low gradeneoplastic proliferation of
(12:57):
cells, then we feel comfortablecalling it DCS. In reality, I
will tell you, myself included,my senior colleagues in
pathology, we actually allow formore. We're pushing that border.
We're moving that bar.
We allow for three millimeters,maybe four millimeters. We like
to quote see more, and there'snot really a a good definition
(13:17):
of more look what more lookslike, but we are pushing that
bar. We're being safer. We arenoticing that with all the
advancement that's happened inthe past, again, forty, fifty
years in breast medicine, breastcancer mortality increases. So
we've been excellent in catchingsmaller and smaller lesions.
(13:39):
We're talking about couplemillimeters. Our radiology
colleagues are phenomenal. Theycatch small calcifications,
small lesions. We excise them.We treat the patient, and yet
there's no really effect onmortality from invasive breast
cancer.
And that's really what we'retrying to do. We're trying to
prevent breast cancer bytreating ADH, atypical ductal
hyperplasia, and low grade DCSductal carcinoma in situ, the
(14:03):
nonobligate precursor. Right? Somaybe we're overtreating or
maybe we're overdiagnosing theselesions and being a little too
aggressive. So back to don't bea hero.
You're doing service to yourselfand the patient by being careful
and admitting challenges. Andthat I would also go on a little
(14:23):
bit of a side note here andmention when you admit
challenges, when you have acommon that explains, hey, this
is an in between lesion, this isa borderline, a threshold
lesion, you prevent themisconception that another
pathologist is wrong. Becauseoftentimes what patients do, and
they're extremely smart, theywill go to another institution,
(14:46):
they will seek a second opinion.And depending on how the other
pathologist is trained,remember, going back to this is
just an interpretation, dependson the person interpreting it.
Right?
The pathologist. Depending howthat person is trained in their
personal approach andunderstanding of the literature,
they may be just a little moreaggressive. And then two
different diagnosis may happen.And then the patients may think,
(15:08):
well, you know, one pathologistis wrong. Which one is it?
These are two separate lesions.And there's a lot more
understanding than that that wedon't convey, that these are
related lesions from the lowgrade neoplasia pathway and that
they're threshold lesions.Right? So by being careful and
not being a hero, you're helpingyour colleagues and prevent the
misconception of somebody, apathologist, being wrong.
Justin Kreuter, M.D. (15:31):
I really appreciate that. And I think it
underscores for the clinicianlisteners' podcast, right? If
you see a pathologist that'susing a descriptive diagnosis,
maybe they're using a diagnosisthat you're unfamiliar with or
like doctor Komforti mentioned,basically using see comment in
the diagnosis field. I thinkit's highlighting that
(15:53):
definitely see the comment andalso don't hesitate to reach out
with a phone call.
Miglena Komforti, D.O. (15:59):
Yeah. That's right. When we write
these reports, we write them fora wide audience. So, obviously,
the patient, obviously, thesurgeon, radiologist, but
there's billing. There'sresearch.
There's the other pathologistwho, again, if the case goes
out, the material is sent toanother institution. There's the
trainee. I talk to myselfsometimes. I use certain
(16:21):
terminology and certainsentences. So in a situation
where I get a phone call,anybody asking about my report,
I can tell you exactly what Isaw because I understand what my
words mean to me.
We speak to a wide range ofpeople, the nurses or the techs
who code these lesions. Right?They have to have a drop down
menu. They have to codeeverything. It's very
(16:42):
complicated.
So every so often, things willbe conveyed in a way that's
perhaps suboptimal because it'snot a diagnostic bucket, and we
like buckets because they followan algorithm, a treatment
algorithm. But we also have toadmit that breast pathology has
those threshold diagnosis andlesions, and it's just by
(17:06):
playing the numbers. Eventually,you'll be faced with one of
them.
Justin Kreuter, M.D. (17:09):
And I appreciate you've taken us
through a little bit of, youknow, how to work through these
challenges. Kind of putting onmy looking into the future hat,
what do you think? I mean, isthis something that, you know, I
think you mentioned, right, it'sbeen several decades since the
criteria was previously revised.I mean, is this example that
you've mentioned today, thisductal carcinoma in situ versus
(17:32):
atypical ductal hyperplasia, isthis threshold diagnosis just
going to continue to be adiagnostic challenge or there's
gonna be new technologies or newways that we may separate these
more reliably, and maybethere'll be different challenges
going forward in the future.What do you see on your future
(17:55):
prognosticating for how how thisis gonna look?
Miglena Komforti, D.O. (17:58):
I think it's very exciting. You know,
some of my colleagues who arelistening or for the students
and residents who are thinkingbreast pathology, they may be
discouraged because they'rethinking, oh my god. This is so
hard. And it is hard, but it'sexciting because there's so much
more work to be done, so muchmore to be discovered. It's
truly, we don't know, right?
And you could be the next personwho cures cancer, I don't know.
(18:22):
But I do think there is going tobe a shift, and I think part of
that is AI. We've seen AI helpus on many different things and
fronts, in breast pathologyincluded in Mayo Clinic Florida.
We do automated interpretationwith AI for some of the
biomarkers in breast cancer, forexample, proliferation index,
MIB. In Rochester, they doestrogen receptor, progesterone
(18:44):
receptor HER2.
So it's extremely helpfulbecause it provides that
precision. It is not asubjective, it is an objective
tool. It's reproducible, and itprovides precision. I think
there's going to be animprovement in the way we render
diagnosis just by incorporatingsome of those AI tools. But I
(19:05):
also think that there has to bean added layer, perhaps
incorporating molecular into ourdiagnosis and not just
immunophenotypic findings andmorphology.
I think there has to be an addedlayer, and I know there are a
couple of companies out thereworking on that. And hopefully
in a couple of years, if we dothis again, I may have some more
(19:26):
information for you that, youknow, we can share. But as of
right now, work in progress.
Justin Kreuter, M.D. (19:32):
Right on. We'll definitely be bringing you
back. One final question I havefor you since you brought up and
we're talking about AI. As apathologist, I work a lot with
the medical school. Mhmm.
And I think that pathology isone of these careers that I
think there's a perception amongmedical students of, in the
world of AI, you know, what is apathologist going to do? What
(19:55):
are your thoughts as an atomicpathologist in AI? You know,
your answer there, youhighlighted, was really talking
about it kind of almost behavingas an adjunct of kind of
integrating additionalinformation, then you as the
pathologist overseeing that. Butwhat do you think that future
interface is going to look like?
Miglena Komforti, D.O. (20:16):
Right. This is a tale as old as time.
So it I've been told by mysenior colleagues that when
immunohistochemistry came out,people were pathologists were
like, oh my god. This isthreatening our jobs. And then
molecular came out, andpathologists were like, oh my
god.
It's threatening our jobs. So inother words, this will be the
(20:37):
solution to our problems, and itwas not. And I kind of hear a
little bit of that with AI whereAI is out and it's going to
solve all of our problems andpathologists will be out of
jobs. Right? Something thatpeople forget is that it is the
breast pathologist, and becausewe're talking about breast
pathology, but it's the breastpathologist who teaches that AI.
(21:00):
So who builds the ground truthfor the AI? It is that
pathologist. And I've had acouple of projects where we sit
down and we annotate and weteach the algorithm what cancer
looks like, what benign lookslike. So there has to be a human
element. There has to be a humanelement.
And I think we need a lot ofprospective data before we can
(21:22):
remove the human element. Untilthen, I think we'll be okay.
Justin Kreuter, M.D. (21:28):
You know, I I totally agree with you and I
see that that role and and Ithink my own thought on this is
you're highlighting this abilitythat it's going to enable you to
focus on even other thingsmaybe. And and as you pointed
out in in the specific case ofbreast cancer, maybe what you
are able to focus on goingforward is gonna be something
(21:48):
that moves the needle as far aswhat does the mortality look
like for men and women whosuffer from breast cancer.
Miglena Komforti, D.O. (21:55):
That's right. And there's a lot to do
in pathology, and I urge you toexplore it as students or or
residents interested in thatspecialty. It's not just sitting
in your office looking at slidesall day. It's very exciting.
There's a lot of, like wementioned, AI.
There's education. There'sresearch, trials, innovation. It
is just a very exciting time.
Justin Kreuter, M.D. (22:17):
We've been rounding with Dr. Komforit
talking about navigatingthreshold diagnoses of the
breast. I really appreciate youjoining us today.
Miglena Komforti, D.O. (22:25):
Thank you. My pleasure.
Justin Kreuter, M.D. (22:26):
And to our listeners, thank you for joining
us today. We invite you to shareyour thoughts and suggestions
via email tomcleducation@mayo.edu. If you've
enjoyed this podcast, pleasesubscribe. And until our next
rounds together, we encourageyou to continue to connect lab
medicine and the clinicalpractice through educational
(22:47):
conversations.
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.
24/7 News: The Latest
The latest news in 4 minutes updated every hour, every day.
The Joe Rogan Experience
The official podcast of comedian Joe Rogan.