Prostate Cancer Diagnosis and Prognosis: Histopathology

September 21, 2021

Information

September 19, 2021

Yale Cancer Center

visit: http://www.yalecancercenter.org

email: canceranswers@yale.edu

call: 203-785-4095

ID6911

To CiteDCA Citation Guide

00:00Funding for Yale Cancer Answers
00:02is provided by Smilow Cancer
00:04Hospital and AstraZeneca.
00:08Welcome to Yale Cancer Answers with
00:10your host doctor Anees Chagpar.
00:12Yale Cancer Answers features the
00:14latest information on cancer care by
00:16welcoming oncologists and specialists
00:18who are on the forefront of the
00:20battle to fight cancer. This week,
00:22it's a conversation about prostate
00:24cancer with Doctor Peter Humphrey.
00:26Doctor Humphrey is a professor of
00:27pathology at Yale School of Medicine,
00:30where Doctor Chagpar is a professor
00:33of surgical oncology.
00:35Peter, maybe we can start off by you
00:36telling us a little bit about
00:38yourself and what you do.
00:40I'm a practicing surgical pathologist
00:42which basically means that I
00:45look at a glass slide under a
00:48microscope and render a diagnosis,
00:50often cancer diagnosis on tissues that
00:54we received from other physicians,
00:57including biopsies and and resections,
01:02and so microscopes have always interested me.
01:05Ever since I was very young and
01:08looked through a microscope at pond
01:10water when I was in elementary school,
01:13but it was actually taking
01:16care of a patient in medical school
01:18that really helped direct me into
01:21pathology and if I can give that story.
01:25I was a third year
01:29medical student and
01:29hadn't really decided on a specialty.
01:33And was considering a number
01:35of different specialties,
01:36including medicine and internal medicine.
01:38Pathology wasn't so
01:41high on the list until
01:45there was an occurrence with one
01:48patient and she was on the internal
01:51medicine ward and she had rib pain
01:55and the radiologists were able to
01:58identify a lesion in the rib and
02:01the differential diagnosis that
02:03we considered clinically and the
02:06radiologist considered was quite lengthy
02:09so it really took a biopsy which
02:13then went to surgical pathology and
02:16in order to establish the diagnosis,
02:18and so I I went down to surgical pathology.
02:21the laboratory where the attending
02:24surgical pathologist was looking,
02:26at slides with their resident
02:28on the service and I asked if they
02:31had seen the biopsy from this
02:33particular patient and he said he had,
02:36and then he pulled out the slide
02:39and went through it and in pretty
02:41short order said oh,
02:43this is metastatic cancer from
02:45the salivary gland,
02:46which was a diagnosis that was not
02:49really considered in this patient.
02:51It turns out she did have a history of
02:54salivary gland cancer 10 years prior
02:58so it occurred to me that this was
03:01the way to really help patients
03:04by helping render diagnosis.
03:08I find that fascinating, because certainly
03:10if you had a patient with rib pain,
03:13metastatic cancer from a salivary
03:15gland would not be top of the list.
03:18Did the pathologist know about the distant
03:21diagnosis of salivary gland cancer?
03:24I think this particular
03:26cancer was so distinctive that
03:28he was able to suspect salivary
03:31gland cancer right away.
03:33I'm not sure if he knew the history,
03:35but he was an excellent
03:37surgical pathologist and being
03:38an excellent surgical pathologist
03:40I'm sure he had asked the
03:42resident for the history first,
03:43as they examined the slide.
03:46Yeah, it's just absolutely fascinating,
03:50but now you've kind of transitioned
03:52still looking at cancers,
03:54but now you're into the world
03:56of genitourinary pathology,
03:58tell us a little bit more about how
04:01your interests transitioned to that.
04:07In residency a big
04:10part of pathology residency,
04:12which is pretty broad based,
04:13we rotate through a number of different
04:17services, subspecialty services and
04:20those services work with specific
04:22clinicians and it's disease
04:25focused and usually organ site
04:29focused. For example,
04:32as a genitourinary pathologist,
04:34I interact very closely with the
04:38urologist and medical oncologist
04:40to treat urological cancers as well
04:43as radiologist and interventional
04:45radiologists to deal with these
04:47type of cancers and specifically
04:49for the genitourinary system,
04:51this is just an introduction.
04:53We basically address cancers that arise in
04:57the prostate and testis and bladder and
05:01kidney, so it turns out when you are
05:06in formative years,
05:08one should never underestimate
05:10how a single patient or a physician
05:14can impact the development of
05:18the individuals who are young and
05:20deciding in medical school or pathology.
05:25So I was a first year resident and
05:28I rotated through the VA hospital
05:31which was right across from Duke
05:33University Hospital,
05:34which is where I did my residency.
05:37And there was a fascinating rotation,
05:39and another excellent surgical
05:41pathologist was the attending there,
05:44and we saw quite a lot of prostate cancer.
05:48And at the VA hospital,
05:51this was several decades ago,
05:55dating myself a number of decades ago,
05:58there was not much known about
06:00prostate cancer,
06:01and treatments were relatively limited,
06:04so it seemed to me that this was
06:06an area where
06:07there is much to be learned about diagnosis
06:10and prognosis as well as treatment of
06:13that particular cancer.
06:15So that's really how I became
06:17interested as a first year
06:19pathology resident in
06:21genitourinary cancers,
06:22and specifically prostate cancer
06:24Let's dive a little
06:26bit more into prostate cancer.
06:28I think that so much of again,
06:31what we do is really dictated by
06:34the biopsies that we take.
06:36So if somebody has
06:38a mass in the prostate or an enlarged
06:42prostate, even more globally,
06:43sometimes a biopsy will be done,
06:46and that'll be sent to the
06:48pathologist and it's really up to
06:50you to try to figure out is this
06:53cancer or is this something benign?
06:56And if it's cancer,
06:57how bad of a cancer is it which
07:00really dictates
07:02is this something that we treat at all,
07:05or something that we simply watch?
07:09How do you make those decisions?
07:11How do you make that differentiation from
07:15benign to malignant and within malignant,
07:18the different grades of prostate cancer?
07:21So it's really quite a long
07:24educational process to be able to
07:27diagnose benign versus malignant,
07:29and it turns out that what's so fascinating
07:31is that every single biopsy is different,
07:34even if we render an
07:36umbrella diagnosis of benign tissue.
07:38For example, the lining tissue of the prostate.
07:41There could be a number of
07:43benign mimicker's in there,
07:45meaning benign tissue looking
07:46like cancer under the microscope,
07:49but it's not, and we have
07:55a differential diagnosis.
07:56We consider a number of different
07:57benign entities before deciding
07:59on a malignant diagnosis,
08:00because that's such a huge step
08:02to take for us and for the patient
08:05and the patient's treating physician.
08:09So that's what I particularly enjoy,
08:14that diagnostic work and it
08:16can be arduous sometimes.
08:18Sometimes it's very straightforward
08:20that a particular biopsy is benign
08:23and sometimes straightforward that
08:25it's malignant, but other times
08:27there are benign conditions under
08:30the microscope that look like cancer
08:32and cancer that can look benign.
08:34So we've been fortunate in this area
08:37to have some tools to help us and
08:40those include antibodies that can
08:42help us recognize specific cells under
08:45the microscope and in certain cases
08:47that can be relayed to us,
08:50but it still requires judgment and
08:53having formed a differential diagnosis
08:56or consideration of what's possible
08:58before we order those tests.
09:00On that issue,
09:02so once having established a
09:04diagnosis of malignancy,
09:06then the next step is to decide and this
09:09is so important for prostate cancer,
09:11how aggressive is it?
09:13Because it turns out most men who
09:16have prostate cancer will die
09:18with it rather than of it.
09:20So there are a large number of
09:23prostate cancers that can grow
09:25very slowly and may not affect
09:27the man during his lifetime,
09:29yet it turns out that
09:31prostate cancer is the second most
09:34lethal cancer amongst American men
09:37by total numbers
09:38trailing only lung cancer.
09:41So those are the cancers we want
09:43to specifically separate out from
09:45the more slowly growing ones.
09:47And we do that under the microscope
09:49using a very powerful approach
09:52that is grade, as you suggest.
09:55So what is grade?
09:57It's basically the way the cells
09:59grow within the prostate once
10:01we've identified them as cancer cells,
10:04so we can look under the microscope
10:06and in their patterns
10:07there are specific patterns that
10:10are known to correlate with the
10:13outcome for the patient,
10:14and so we have various tiers,
10:17various numbers we can apply
10:20and the most simple one that we
10:22use right now is grade group and
10:24that ranges from one to five.
10:26One being the best outcome,
10:29and those patients are managed
10:31very differently from
10:32those who have a grade group
10:35five out of five,
10:36but there's everything in between,
10:38so it's really a spectrum.
10:39And therein lies again
10:41judgment as far as deciphering as you note,
10:46the detective work deciphering
10:48out the patterns that can help
10:51us assign a grade that
10:53indicates aggressiveness of
10:55that prostate cancer.
10:58One of the questions
11:00that I think always comes up is
11:02that it seems to be a little bit
11:04of art and a little bit of science.
11:06Looking at these patterns
11:09and trying to decipher is this
11:12lower grade?
11:13Is this a higher grade?
11:15How much of it is art,
11:18and how much of it is science
11:21and how sure are you at any given
11:24time of your diagnosis being correct?
11:27Interpretation of slides
11:30under the microscope is most definitely
11:34both art and science, so there's
11:37much experience that one must have
11:39in order to recognize these patterns.
11:42The science part is that we can use
11:45antibodies to help identify specific cells.
11:48The grading part remains, though,
11:50very much art and pattern recognition
11:53going forward in the future,
11:54and this has already started.
11:56We have tools that can help us recognize
11:59patterns even better and more quantitatively,
12:02and that's through the use of artificial
12:06intelligence and machine learning.
12:08So all of that work has just started,
12:10but already I've had the opportunity
12:12to work in on a couple different
12:15projects and it turns out that the
12:18computer with specific algorithms
12:20can identify,
12:21can diagnose and grade prostate
12:23cancers just as well as a number
12:26of us who specialize in sub
12:28specializing in that particular area.
12:31I can't wait to learn more about that,
12:33but first we need to take a
12:35short break for medical minute.
12:37Please stay tuned to learn more about
12:40prostate cancer diagnosis and prognosis
12:41with my guest doctor Peter Humphrey.
12:44Funding for Yale Cancer Answers
12:46comes from AstraZeneca, dedicated
12:48to advancing options and providing
12:50hope for people living with cancer.
12:53More information at
12:56astrazeneca-us.com.
12:58The American Cancer Society
13:00estimates that nearly 150,000 people
13:02in the US will be diagnosed with
13:05colorectal cancer this year alone.
13:07When detected early, colorectal
13:09cancer is easily treated and highly
13:11curable and men and women over
13:13the age of 45 should have regular
13:15colonoscopies to screen for the disease.
13:18Patients with colorectal cancer
13:19have more hope than ever before,
13:21thanks to increased access to advanced
13:24therapies and specialized care.
13:26Clinical trials are currently
13:28underway at federally designated
13:30Comprehensive Cancer Centers.
13:31Such as Yale Cancer Center and
13:33at Smilow Cancer Hospital to
13:35test innovative new treatments
13:37for colorectal cancer. Tumor
13:39gene analysis has helped improve
13:41management of colorectal cancer
13:43by identifying the patients most
13:45likely to benefit from chemotherapy
13:48and newer targeted agents,
13:49resulting in more patient specific treatment.
13:52More information is available at
13:55yalecancercenter.org. You're listening
13:57to Connecticut Public Radio.
13:59Welcome back to Yale Cancer Answers.
14:02This is doctor Anees Chagpar and I'm joined
14:04tonight by my guest doctor Peter Humphrey.
14:07We're talking about prostate
14:08cancer diagnosis and prognosis,
14:10and right before the break we were
14:13talking about this magic that
14:16happens in the pathology lab.
14:18At least, it seems like magic to those of
14:20us who send them biopsies and magically
14:23get back a diagnosis that we then
14:25use to treat our patients and Doctor
14:27Humphrey was telling us that this is
14:29in part art,
14:30but it is in part science and
14:33that you're able to use antibodies
14:36and so on to help you in making
14:39that diagnosis and right
14:42before the break you started to talk
14:45Doctor Humphrey about artificial
14:47intelligence and how this might actually
14:51help us in making a diagnosis now,
14:55so that the computers might
14:57be able to make a diagnosis
14:59almost as well as an experienced pathologist.
15:02Tell us a little bit more about that.
15:05So we are in the very early pilot stage
15:09I would say as far as
15:12development of this tool,
15:14but I think it will be an important
15:16tool that can assist the pathologist
15:18and actually artificial intelligence
15:20is being developed in many
15:22branches of medicine and
15:24radiology too, so it turns out
15:26that those parts of medicine that
15:28deal with diagnostic images like
15:31radiology and pathology are areas
15:33where there could be great benefit.
15:35From more standardization, I would say,
15:39and perhaps even quantitation,
15:42using computer assisted methods,
15:45so that's already happening and
15:46actually happening very quickly
15:48as far as the research into this.
15:50And the use of computers and
15:53artificial intelligence.
15:55To develop algorithms, ways in which
15:59the computer can diagnose and
16:02even grade prostate cancer.
16:05So I've been fortunate enough to have
16:07been involved in a couple of these
16:09research studies and a number of us
16:11from around the world who
16:13are interested in prostate cancer
16:15and are genitourinary pathologists,
16:17and we've looked at hundreds of slides,
16:19all online,
16:20so these are all images diagnosable
16:23on our computer.
16:24And then we tested the algorithm and then
16:30the algorithm was tested against our
16:33diagnosis in grade versus collections
16:35of pathologists who were
16:39not sub specialized in
16:41prostate cancer diagnosis and the
16:43computer was actually just as good
16:46as our diagnosis and grading.
16:49So what does this mean for the future?
16:52Well, there are actually a lot of challenges.
16:55There are several algorithms that
16:57have already been published.
16:58Methods that the computer uses,
17:01and there's a lot of standardization
17:04and validation that needs to occur
17:06so that a computer can use
17:08images from a particular laboratory
17:11and one particular hospital as far
17:14as the scanners they use to make
17:16those images and the way the slides
17:19are prepared that all of those
17:21factors can have a huge impact on the
17:23success or failure of the algorithms.
17:26My hope is that as far as standardization,
17:29it can be used as a tool to help
17:31hospitals where there may not be ready
17:34access to a genitourinary pathologist.
17:37And also I think for those of
17:38us who have high volumes
17:40and have a special sub specialized group
17:42of Geo pathologist as we do here at Yale,
17:45I think it might actually
17:47help us screen cases.
17:48So that the computer could actually help
17:51us identify through all these slides,
17:54identify the ones that need particular
17:57attention or standardized grading.
17:59So there may be,
18:00for example,
18:01a difference of opinion about the
18:03grade of a specific cancer and
18:06the way we currently address this,
18:07and this is very important actually
18:09when there's a difficult case,
18:10we'll have a consensus conference,
18:13meaning that up to seven of us
18:15who are sub specialized in
18:18genitourinary pathology at Yale will meet
18:19around the microscope or in this area,
18:22from our computers and look
18:24at the images together to try
18:26to agree on a particular grade.
18:28In a difficult case or where it's a
18:30borderline case between grades for example.
18:32So maybe the computer could also
18:35provide help in standardizing those.
18:38Those sorts of assessments when
18:40it's a difficult or borderline case,
18:43so it sounds like this is really exciting
18:47technology that might be able to provide
18:50a second opinion. But for right now,
18:53if you're a patient and you might not
18:57be at or near a large academic center,
19:00and you get a prostate biopsy,
19:03for example, how important is it for
19:05you to get a second opinion on that
19:09biopsy from another human pathologist
19:11if a computer isn't readily available?
19:14That's a really critical question,
19:17and I think it's important to
19:19know in discussions with your
19:22physician whether a genitourinary pathologist
19:25has reviewed the slides and it's
19:27true that around the country there
19:29are just varying degrees of practice
19:32and varying volumes of practice,
19:35and so at a smaller hospital,
19:37maybe only a few prostate biopsies might
19:39be seen over a long period of time,
19:42and particularly in those cases where the
19:45pathologists may not feel as comfortable,
19:47or the treating physician may not
19:50feel as as comfortable, it's
19:54I think a useful step to seek
19:56a second opinion,
19:57and we see slides for second opinions
20:00all the time here from everyone.
20:02Actually from patients from treating
20:05physicians and from pathologists themselves,
20:08and this is an important quality
20:10to all these second opinions.
20:12And again we very commonly almost
20:15on a daily basis share cases here at
20:18Yale amongst our group of seven genitourinary
20:21pathologists
20:23And so are these second opinions when
20:26you go and you have your
20:28slides reviewed by somebody else?
20:30Or maybe the pathologists themselves sends
20:33it to another center to get reviewed
20:36if they're not quite sure
20:38about the diagnosis,
20:39is that covered by your insurance?
20:42Usually it is.
20:44So at least the cases that we
20:46receive here for second opinions.
20:50That's good to know.
20:52Is it ever the case where even
20:55if you go to a large academic
20:58center that it's worthwhile
20:59getting your slides reviewed by
21:01another large academic center?
21:03I mean, how much heterogeneity
21:06is there between experienced
21:08genitourinary pathologists for example?
21:11So since diagnosis
21:13and grading are still art,
21:15there can be differences of opinion amongst
21:19even expert and experienced genitourinary
21:22pathologists and these tend to be the
21:25rarer or more borderline cases.
21:28There's been a lot of research looking at
21:32variations or differences of opinion
21:35between pathologists and even between genitourinary
21:38pathologists,
21:39and even did a study where I looked
21:42at agreement with myself so I
21:45diagnosed and graded some slides and
21:48then came back sometime later
21:50to see if the diagnosis and grading
21:52were the same so the agreement is
21:54pretty good amongst genitourinary pathologists,
21:57but one should not hesitate in
21:59seeking a second opinion at another
22:02center with an established
22:04group of pathologists,
22:06but as we
22:08talk about that variability,
22:09all of these pathologists are looking
22:12at the same slides and I know that in
22:15other cancers we've talked on this show
22:17about this concept of
22:19heterogeneity that you might have a
22:22cancer that looks kind of different in
22:24one part than another and so I wonder,
22:28when you get these biopsies,
22:30we often
22:33send a core biopsy so
22:35a sampling of this tumor,
22:37how representative is that,
22:38and is it ever the case where
22:41you look at this and you
22:43kind of say
22:45I don't know that this is representative?
22:47We need to get more tissue or
22:49are you usually pretty happy
22:51with the sample that you get?
22:54So that's such a key question and
22:57really the practice of the biopsy
23:00as far as the prostate has changed so
23:04remarkably since when I was a resident.
23:06So back in the olden days,
23:08it was usually just one needle biopsy,
23:11digitally directed towards a palpable
23:13mass in the prostate by the examining
23:16physician and one single core was taken.
23:19So prostate cancer,
23:20heterogeneous concept of heterogeneity,
23:23and different areas of the prostate
23:25being actually of different grades
23:27and different aggressiveness
23:29is actually characteristic of
23:31prostate cancer and prostate
23:33cancer also tends to have multiple
23:35nodules within the same gland,
23:37so what's been a real advantage
23:39is medical advances in radiology,
23:41and there are expert radiologists here who have
23:45actually helped develop this technique,
23:47and that's a special type of MRI.
23:50Magnetic resonance imaging that's used
23:53with ultrasound to guide the
23:58needle placement within the prostate.
24:00So now rather than one needle core,
24:03we often receive anywhere from 20 to even
24:0830 individual needle cores per patient.
24:12And the reason is that the radiologist
24:14now can identify areas where they're
24:17suspicious of cancer and can specifically
24:20say based on their grading scheme,
24:22whether they think it's a lower
24:25risk or a higher risk case so
24:28I do feel good about the
24:32representation for most patients and when
24:35the patients have undergone this
24:38type of imaging by the radiologists.
24:43Even though multiple needle cores
24:44are placed in a single nodule,
24:46it's still possible that maybe
24:49a smaller high grade area was missed.
24:53Warning signs would be,
24:54what if the patient has a really high
24:57serum PSA prostate specific antigen level?
25:00Or what if this is radiologically
25:02a very aggressive looking lesion,
25:04but we don't see that under the microscope?
25:07Then I would worry about
25:10the needle maybe not sampling
25:12the worst of the cancer.
25:14Yeah, it goes back to that
25:15concept of being a bit of a
25:18detective that we talked about before
25:20the break and the fact that the
25:22pathologist is really a key part
25:25of this multidisciplinary team that
25:26you need to get information from.
25:29From the radiologist,
25:30from the surgeon.
25:32from the other physicians.
25:33who are involved
25:34in the case to kind of put all
25:37of the pieces together to make
25:38sure that it all makes sense.
25:40That's what I love about
25:43working here is working with so many
25:45bright and experienced physicians
25:47who are passionate about providing the
25:50highest level care and talking with
25:53them about what their perspective
25:56and view is on a specific patient.
25:59For example, if there
26:02is not a link made between pathology
26:06and what we see in the clinical setting
26:09that sort of correlation is
26:12clinicopathologic correlation and
26:13is so vital.
26:16Going back to that patient with pain in the rib.
26:19It was absolutely essential to know that
26:21the patient had a history of cancer
26:2310 years ago to establish firmly that
26:26cancer scene and what we would
26:28do is compare slides.
26:29That cancer in the rib biopsy
26:32was the same as the cancer in the
26:35salivary gland,
26:36so we do that commonly to look
26:38back at old slides to see if
26:40if cancer has come back and
26:42we think a cancer might
26:45have come back or spread so that
26:47comparison is a really important part
26:49of the detective work we do.
26:52And I think the other piece
26:54that's so important is that it's so
26:57critical in terms of what you do,
26:59especially in prostate cancer,
27:01to really nail down how
27:03aggressive this is because it is
27:05the difference these days between having
27:08more aggressive surgery
27:11or radiation versus watchful waiting.
27:14Tell us a little bit more about how your
27:18decisions impact treatment and prognosis?
27:21After establishing a
27:23diagnosis of prostate cancer,
27:25we assign the Gleason grade or score
27:28and that is a grade number we give
27:31for every single prostate cancer
27:34needle core in every case and a great
27:38group for that particular biopsy.
27:41So if a patient had 10 positive
27:43cores with cancer in each one,
27:45we would assign an individual
27:47grade to each one,
27:49and actually I just gave a
27:51lecture this morning to the pathology
27:53residents on grading and staging.
27:55So it is one of the most critical
27:59things we do because grade is such
28:02a dominant prognostic indicator for us.
28:06For the patients physician,
28:08for everyone,
28:09and the patient themselves.
28:13For example, a grade Group One in a patient
28:15with a lower PSA might consider
28:18along with their physician,
28:20the physician might consider active
28:23surveillance or careful monitoring of
28:25that cancer compared to a grade Group
28:275 where everyone would agree this
28:30patient definitely needs active therapy.
28:33Doctor Peter Humphrey is a professor of
28:35pathology at the Yale School of Medicine.
28:38If you have questions,
28:39the address is cancer answers at
28:41yale.edu and past editions of the
28:44program are available in audio and
28:46written form at Yale Cancer Center Org.
28:48We hope you'll join us next week to
28:50learn more about the fight against
28:52cancer here on Connecticut Public
28:54radio funding for Yale Cancer
28:56Answers is provided by Smilow
28:57Cancer Hospital and AstraZeneca.