A pathologist explains prostate cancer diagnoses, including staging and grading
Host Amber Smith: Upstate Medical University in Syracuse New York invites you to be The Informed Patient with the podcast that features experts from Central New York's only academic medical center. I'm your host, Amber Smith.
Once a man learns he has prostate cancer, before he and his doctor can decide on treatment, they'll need information about the growth or spread of the cancer.
That's where laboratory pathologists come in. Today, I'm talking with pathologist Gustavo de la Roza, a professor and vice chair of pathology at Upstate, and also the director of anatomic pathology. Welcome to "The Informed Patient," Dr. de la Roza.
Gustavo de la Roza, MD: Thank you. I'm glad to be here.
Host Amber Smith: The American Cancer Society recommends screening for prostate cancer using the prostate specific antigen, or PSA, starting in a man's 40s or 50s, depending on his individual risk.
This is the blood test, but can you explain to us what the test is looking for?
Gustavo de la Roza, MD: Prostate specific antigen, it's normal in the body, so It's an enzyme it's in the prostate, but it gets released into the bloodstream as well. So, when there's pathology in the prostate gland, this antigen could be increased in the serum, and that's what we measure.
Host Amber Smith: So what does the number mean then? Because a man would get a number back from the lab, right?
Gustavo de la Roza, MD: It depends a little bit on the patient's age, but in general, I would say, for practical purposes, any level greater than 4 is considered abnormal. So it's 4 -- it's nanograms per milliliter -- but basically 4. If you say that, people would understand.
Host Amber Smith: Is it normal for the number to fluctuate up and down from one PSA test to another?
Gustavo de la Roza, MD: Yes. And also, it's important to mention that the PSA is not specific of cancer, so it's elevated in a lot of processes that happen in the prostate. The most common one is what people know as BPH, or benign prostatic hyperplasia. It's almost, if you will, a physiological change because all men get it. It enlarges the prostate and so forth. And so that increases also the level of serum PSA. And so does prostatitis, so infections or inflammation of the prostate.
Host Amber Smith: So just because you have a high number, higher than 4, doesn't at all mean that it's ...
Gustavo de la Roza, MD: Exactly. No, no. In fact, it's been questioned, a little bit, it's use as a screening test. And the reason why is because it was used so much and patients were biopsied and diagnosed with cancer that probably doesn't need to be treated, it resulted in a lot of overdiagnosis and overtreatment. But that's kind of another subject.
Advanced prostatic cancer is uncommon, uncommonly diagnosed. So in other words, if we didn't have prostatic specific antigen, we'd be missing a lot of early cancers, and we would only detect those that are already advanced.
Host Amber Smith: Well, let's talk more about a biopsy, and that's where you would, get a tissue sample from the prostate. Is that the only way to get a definitive diagnosis of prostate cancer?
Gustavo de la Roza, MD: Yes. Yes, it is. Clinically, it's required for any treatment. The truth is, if you have a widespread metastatic disease in the bones and the PSA is super high, the likelihood that that is cancer is very, very high.
Regardless of that, confirmation by a tissue sample is needed.
Host Amber Smith: In the lab, what are you looking for in those tissue samples?
Gustavo de la Roza, MD: What we're looking for is a proliferation of abnormal glands. So they're glands of the prostate that resemble the normal glands, but they're different under the microscope.
Host Amber Smith: What do cancer cells look like?
Gustavo de la Roza, MD: In prostate cancer, basically, the cells are larger and they grow in a disordered way, so in a kind of haphazard pattern. So that, and there're also very cytological features, meaning precise features of the gland, the cells itself. All the cells have a nucleus and a cytoplasm, so features of that nature that would make us, diagnose cancer.
And they're different from the normal glands. And there're also ways of doing some markers that we can use, immunocytochemistry, which is a form of tumor markers that will be present in benign tissue and not in cancer.
Host Amber Smith: Are you able to tell, based on the biopsy sample, how advanced the cancer is, or can you tell whether it has likely spread?
Gustavo de la Roza, MD: Not really. There is a component of that, but it's very uncommon that if you see in a biopsy that there's a little bit of adipose tissue, or fat, and there's tumor in it, we know it's spread out, but that's very uncommon. So in general, no. What you can use the biopsy (for) in general, is a way of predicting which cancers could be already outside of the prostate.
Meaning if you have a very high-grade tumor, the likelihood of the cancer have been already out of the prostate is much higher.
Host Amber Smith: How common are false negative results in biopsies, where a patient learns that they do not have cancer, but actually it turns out that they do. Does that ever happen?
Gustavo de la Roza, MD: Yeah. There is roughly, about probably 2 to 2.5% with today's protocols. That was not an uncommon event a couple of decades ago, or even a decade ago, maybe, where the biopsy sample was more restricted, and the number of biopsies were lower, and so we really missed a lot of cancers.
But today, that's a lot lower possibility.
Host Amber Smith: What are the possible reasons for a false negative at this point in time?
Gustavo de la Roza, MD: In general, the concept is that more biopsies lead to a higher detection rate. So it's a think, it's a matter of sampling. Did we get it all? Do we have all the areas sampled?
So, just to give you an example and put this in perspective, decades ago, the typical protocol for urologists to perform a biopsy was doing six biopsies,so we call it sextant biopsy. It was a classic for decades and decades, and so usually we divided the prostate in the apex, the mid-prostate and the base of the prostate, and they would do one biopsy from each side, right and left.
Today, the protocols at least include what we call extended protocols, 12 biopsies, which include the periphery of the prostate. That was not included in the previous ones. And the cancers tend to be in the periphery of the gland, so there's a higher likelihood that it will be detected.
In addition to that, in some places they do something that's called saturation biopsies, many of them, actually much more 20 or 30. but that's not used very often. What people may be start seeing now. it's done Upstate, but not in every medical center, is image guided biopsies, also known as fusion biopsies. And they're called fusion because it's a way of targeting a lesion rather than just sampling blindly a lesion. And that's fusion because it's the fusion between ultrasound and MRI testing. So they do an MRI and ultrasound and detect the more likely areas of a lesion. And that is sampled, and that yields a much higher detection of prostate cancer.
It is also worth mentioning that in the past there were some patients that, because of their age, maybe their life expectancy is not that low, or the cancer involvement is very minimal. They will be put in (the category of) basically deciding not to treat, and they called it watchful waiting. That was probably about 10 years ago, they started changing that to something called active surveillance. It means the same a little bit, but it's not just waiting for something to happen. You are actually more active in watching the PSA and then deciding to biopsy again. And that is critical. For instance, at Upstate, and in many medical centers around the country, people are not put in active surveillance until this type of biopsies are done. So if they're done this way, the likelihood of missing cancer is much lower, so they feel confident. Whereas if somebody comes with a classic (case), 12 biopsies are negative and the PSA is really high, it makes you wonder whether there's cancer that could have been missed. And so it's safer to put him in active surveillance this way.
Host Amber Smith: You're listening to Upstate's "The Informed Patient" podcast. I'm your host, Amber Smith, and I'm talking with Dr. Gustavo de la Roza. He's a professor and vice chair of pathology at Upstate and the director of anatomic pathology, and today we're focused on prostate cancer.
What can you tell us about how prostate cancers are graded using the Gleason scale?
Gustavo de la Roza, MD: It seems, kind of complicated, but relatively speaking, it's rather simple. So Gleason grading is a form of histologic grading, meaning the least aggressive to more aggressive types, depending on a score. And the truth is, Gleason grading is actually called, in reality, score. So it's a combination. What we're looking at is basically whether the glands are well-formed glands, there're glands that they're not so well formed and they're coalescing together or they're infiltrated in a single file, different things that we use. And the division grade actually has a very good reputation because it's one of the best, probably, grading systems that exists I would say the most reliable.
It was created in the 1960s by a Dr. Gleason. He was in Minnesota at a VA (hospital) in Minnesota. So the grading has been changed a lot. There have been a lot of updates. So what is used in the Gleason grading for people to understand is that we divide it in five different histologic patterns.
And so in the Gleason grading, we use the most predominant pattern. Followed by the second most predominant pattern. So that means if you have a Gleason score of 6, and usually patients will see it says Gleason score 6, in parenthesis, 3 plus 3. Or sometimes it says Gleason 3 plus 3 equals 6, which seems kind of silly, because it's basic math, but that's the idea.
So if you have a Gleason 7, you could have a Gleason seven, three plus four, or at Gleason seven, four plus three, meaning that the four is more predominant, which is actually, a more aggressive form. that's how we create it.
So the lower the Gleason, the better. And the higher is obviously more aggressive. But there's been some very important changes since that, and I think it has helped researchers, technicians and also patients, probably. It would be easier to understand. So the problem with the Gleason system is, like I mentioned, we have five different patterns.
The truth is that the patterns, for instance, what is called pattern 1 and pattern 2, they're almost nonexistent in real practice. This was the 1960s, the new methods. So we know that those probably are not really cancer. So the reality is that the minimum Gleason pattern that we can see is 3.
So if you have a tumor that is all 3, that means it's going to be 6, 3 plus 3. So if one person looks at that, a patient looks at that and know that the scores, the Gleason score, or Gleason grading, goes from 1 to 10. If you're in 6, well,, you have a pretty significant cancer, right? You're not 1, you're not 2, and you're 6. And the reality is that Gleason score 6 is the lowest possible Gleason grading.
So research studies actually looked at what is called chemical recurrence. So basically after the prostate is taken out, the PSA starts to rise. So that's called a chemical recurrence because the PSA should be normal by then, when they took the prostate with all the tumor. So based on that, it has been divided in five different groups and combinations of Gleason grades.
So that's what patients will start to see and maybe possibly Gleason per se, with that name may be actually not used anymore. What we call this is great grouping, and it's five different groups of combinations. So the group 1, for instance, is a Gleason 6, 3 plus 3, and a Gleason 5, it's a Gleason 9 or 10.
So it's a combination of pattern 4 and 5, 5 and 4 or just 5 and 5. So those are the highest Gleason grading. So this is more accurate because it really reflects more, exactly what's happening and the likelihood of that tumor to spread or to be more aggressive.
And better understood.
So today, you will see a biopsy, for instance, I will diagnose a Gleason adenocarcinoma and say Gleason 6. And then he will say how much of the biopsy was involved in cancer? And then that's followed by a great group. And I say great group 1 of 5.
If you have a very high Gleason grade, so for instance, you have a Gleason grade of 8, you would say Gleason score 8, in parenthesis we'll put 4 plus 4 or 3 plus 5. And in parenthesis then we will say great group 4 of 5. So that's easier to understand it that way.
And also this separation is not totally arbitrary. It's based on risk of having a recurrence or, cancer coming back after surgery.
Host Amber Smith: I understand the grading system has evolved since the 1960s. Does it still rely mostly on the pathologist's impression of what the sample looks like under the microscope?
Gustavo de la Roza, MD: Yeah, absolutely. I'm glad you mentioned that. because actually urologists trust the Gleason grading, a lot and it is a very good system, but there is some sensitivity to it, and there's some inter-observer variability, so it's not perfect, but it's pretty good. So the way we grade things has not changed as much, it's how we report it. And today, cancer protocols, and they talk about great groups rather than Gleason grade. It's a lot easier because it's a combination of different, gradings and possibilities.
Host Amber Smith: But if I understand correctly, the grading only tells part of the story, because you also do something called staging. Can you explain what that is?
Gustavo de la Roza, MD: Right. So staging is different. It's a way of judging how much the tumor has spread, how much of the prostate is involved by cancer. Has it gone outside of the prostate? Has it gone to the lymph node? Has it going to other organs, to the bone? And so forth.
So there's two ways of TNM staging, two different types. One is the clinical, which is what the clinicians do in base of imaging analysis, and. X-rays and MRIs and CT scans and so forth.
And then there's the pathologic staging. So the pathologic staging can only happen after the prostate has been taken out. So we see how much of the prostate has been involved, mostly. Just to put it simply, a very low tumor is usually, let's say, T2. Basically (that) means that it's confined to the prostate; 3 and 4, meaning that they went outside of the prostate into the surrounding tissues, the fat, or It could go into the rectum and things of that nature.
The staging that you're referring to is how do we determine, the most accurate and the one all centers all around the world uses, called TNM. So T for tumor, N for nodes, and M for metastases.
So, the tumor will be what I just mentioned. Is it confined to the prostate? Is just outside of the prostate? How far did it go? And then, are the lymph nodes involved? How many of them are involved? Are they involved or not? And then, are there any metastases outside of the lymph nodes or the prostate? So, distant metastasis or distant spread.
Host Amber Smith: Under the T designation, does the actual physical size of the tumor matter?
Gustavo de la Roza, MD: Not in prostate. In many tumors it does, but in prostate it's actually like what I mentioned to you. T1 is reserved only for those diagnosed and the needle biopsy and their pathologic grading. T2 is the lowest possible one, and that means it's confined to the prostate. T3 is when the tumor has spread either to the fat around it, so outside of the prostate or into the seminal vesicles -- that is another component of the prostate. And then, at 4, it's like I mentioned, it goes into other organs, such as, for instance, the rectum, which is very advanced then.
Host Amber Smith: So TNM staging is not a prediction, it's actually what is going on currently.
Gustavo de la Roza, MD: Exactly, exactly. So it's the involvement of cancer in the body, how far has the spreading occurred, or has it occurred at all?
Host Amber Smith: So the urologist takes this information, the grade and the stage, and that's how they recommend treatment?
Gustavo de la Roza, MD: Absolutely. Yeah. So it's a risk assessment that I'm not going to get into. It's quite complicated, but it actually takes into account to show a very low risk, low risk, intermediate, risk, high risk and very high risk. And it's a combination of staging, grading and level of PSA in the serum. So a combination of those puts a patient in a category which is easy to decide how to treat or what to offer the patient in terms of treatment.
Host Amber Smith: When is molecular testing of the tumor recommended?
Gustavo de la Roza, MD: It's very early on that now we have several ways of trying different tests to try to predict aggressiveness of the tumor that go beyond the grading. But the truth is this has not been adopted or has not been really tried well enough to have a universal, use of them.
So the truth is there are many commercially available molecular testing that the companies have developed, and there's been some studies, and that's what they're trying to predict. So, for instance, a patient with a Gleason 6 for the most part will be told that probably doesn't need to be treated.
The question is, at some times, obviously because we are sampling, we can get a 6, but the truth is maybe somewhere in the gland -- we can't sample it all -- it might be a higher pattern or higher grade. So these molecular tests that target different genes in a combination of genes, difference in each one, may be able to predict which patients with that kind of grading will be progressing.
But in a nutshell, there are many of them, and none of them have universal use recommended for now. There haven't been any significant randomized trials for it.
Host Amber Smith: So it would be very patient specific if the doctor felt like it would be useful or not.
Gustavo de la Roza, MD: Right. and it doesn't hurt to get some of those tests.
But the truth is, actually, at Upstate, mostly we're not using them.
Host Amber Smith: What are the most common places where prostate cancer may spread? You mentioned the rectum. Is that just because of its proximity?
Gustavo de la Roza, MD: Right, right. So, in staging, we're talking about local spread or involvement and then distant spread. So one is, like you mentioned, going into the rectum, which is pretty close to it. So it goes Into that, that's direct extension, and that's what we call a T4 tumor. But if the tumor goes to the lymph node, meaning that it has left the prostate already, it's not just in the confines of the prostate, goes to the lymph node, and then after that it will be going to different, more distant sites.
The most relevant and most typical is the bone -- bone lesions. And, lung also. It could be other organs. It just becomes really serious level of metastasis when it goes to the visceral, organs, like the lung and so forth. But the most common is the bone. Once the tumor has spread to the bones, basically we're dealing with a disease that is not curable at that point. Doesn't mean that it's lethal. It could be treated, but it's not curable.
Host Amber Smith: When you find cancer in the prostate, is there ever a concern that it's a cancer that's spread to the prostate from somewhere else in the body? Does it ever happen that way?
Gustavo de la Roza, MD: Yeah. But it is very, very uncommon. So yes, we can have certain types of tumor that could actually go to the prostate, like lung cancer and so forth, but it's very uncommon.
Host Amber Smith: Aside from the biopsy, are there other tests that might be ordered which would end up in the pathologist's lab?
Gustavo de la Roza, MD: Not really, no, other than PSA and the prostate biopsies and then surgical specimens, if the prostatectomy has been performed. We talk about molecular testing, but we don't have any accepted form of molecular testing at this time.
Host Amber Smith: Do you have any idea how common it is for men to live with prostate cancer, even for years? I mean, we talked about active surveillance. Does that happen? Are there a lot of men that are actively ...?
Gustavo de la Roza, MD: Yeah. Yeah, it is. It is. Some of the great developments that happened have been based on this.
So. I cannot be precise in terms of timing. It could have been probably about two decades or maybe 15 years ago. There'd been studies done originally in Europe. The Swedish were the most reliable studies, but then they were done elsewhere in Europe and the United States.
People, realized that the prostatectomies were overtreating patients because no one was progressing into any disease, and the disease had spread already. It was the same whether they had a prostatectomy or not. So there wasn't a lot of value.
So the truth is, we're probably overtreating cancer, for the same reasons that I explained at the beginning, that you have a Gleason 6, so it's an intermediate grade, it's cancer. Today, we know that Gleason grade 6, for instance, has very low potential for spread. So patients, yes, they can live with cancer and not be operated on and followed for it.
There is a saying that some people die with cancer and others die of cancer. Still, prostate cancer is the most frequently diagnosed tumor in men, and the second cause of mortality. So, it is important, butthere are more cases diagnosed. So for instance, if you consider prostate, the estimated new cases, maybe 20% of all the cancers, whereas lung could be 14%, which is less.
But when you're talking about death, lung causes almost 25 to 30% of mortality and prostate, being more common, only causes 9%, so it's less lethal.
Host Amber Smith: Can you tell me about the use of artificial intelligence in pathology?
Gustavo de la Roza, MD: Yes. This is a very modern thing, and it's based on computer imaging of the same images that we looked at under the microscope, but it's scanned perfectly all around the sample.
These computers are basically taught how to detect things, right? And so, it's a very complex process, but the idea can be applied to the prostate very easily, and in fact, it already exists, being approved for FDA approval for use in the prostate. To make it easy and simple, diagnosing cancer of the prostate is not that complicated. We talked about grading, which could be variable, so that the artificial intelligence targeting of all these biopsies that we mentioned, at times all are negative. Sometimes one or two are positive, so that means that we're looking at 15 biopsies with no tumor in them. And so it's very easy to teach a computer how to detect that. So we will concentrate on the ones that we need to concentrate (on). It doesn't mean that a pathologist doesn't look at it, it just is selecting the areas of concern. And the computers are actually much better than humans for this because they don't get distracted, they don't get biased or interrupted by a telephone call and so forth.
The second thing is the histologic grading that we were discussing. A lot of these, I didn't want to, overwhelm people with details, but it's a percentage of a pattern and the other pattern and how we make decisions. So if you teach a computer how to do this based on information, and these computers have a network that actually allows it to learn a system, the grading will probably be more objective and possibly more accurate. But then again, it won't be a computer diagnosing you and grading your cancer. It will be a pathologist doing it with the help of the computer.
Host Amber Smith: Well, Dr. de la Roza, thank you so much for helping us understand more about prostate cancer.
Gustavo de la Roza, MD: Oh, you're very welcome. My pleasure.
Host Amber Smith: My guest has been pathologist, Dr. Gustavo de Roza, Upstate's vice chair of pathology and director of anatomic pathology.
"The Informed Patient" is a podcast covering health, science, and medicine, brought to you by Upstate Medical University in Syracuse, New York, and produced by Jim Howe. Find our archive of previous episodes at upstate.edu/informed. This is your host, Amber Smith, thanking you for listening.