Prostate Health 2011: Screening Dilemma

MD Anderson Cancer Center
Date: January 2012

>> So our next featured speaker is Dr. Robert Volk and he's a PhD researcher, very much interested in the educational process of prostate cancer dilemma and screening, so he's clearly not here to promote one therapy over another.  It's a very neutral view and we look forward to your presentation.  Thank you.

>> Thank you, John.  Good morning.  I appreciate having some time with you all today.  I'm here to talk about educational tools for prostate cancer screening.  I will admit a little trepidation on my part in that I know we're talking to a lot of people who've gone through prostate cancer screening and also treatment.  But today, I'm gonna talk a little bit about some educational tools that have been developed to help patients make informed decisions about prostate cancer screening.  And so I'll talk about the screening dilemma.  I'll talk about new guidelines around early detection for prostate cancer screening, some patient decision aids and also ask for your questions.  Okay, this slide has been reformatted a little bit, but people may be aware of the screening trial results that were published in New England Journal of Medicine back in 2009 reporting the European trial and a trial done here in the United States.  In fact, if I could ask for just a raise of hands of folks who are familiar with those 2 trials coming up back in 2009.  Okay.  Alright.  These are sort of a--[Laughter].  These are sort of game changers in that they really gave us some good information about the potential benefits of prostate cancer screening in saving lives and in decreasing mortality from prostate cancer.  Again, we had a large trial done here in the US, the Prostate, Lung, Colorectal and Ovarian Cancer Trial.  And then there was a parallel trial done in Europe that actually, if you look at it carefully, was a lot of individual trials that were done in different European countries.  And I'll just very quickly, you know, kinda of get to the bottom line for these trials.  So in the US, there are about 75,000 men that participated and they were randomized to one of two groups.  The intervention group received PSA tests annually for 6 years.  They also had digital rectal exams annually for 4 years and then they were followed thereafter.  Another group of men were randomized to no screening.  As an aside, I want to mention that in the US, the PLCO trial, the men who were assigned to the no screening group, about half of them actually went ahead and got screened anyway.  So we have problems when we look at that data in that we don't really have good comparisons on the US side anyway.  The European trial was a little bit bigger, about 160,000 men were screened and they were screened every 4 years.  So as opposed to being screened annually, they were screened every 4 years over a period of time.  The US trial did not find a difference between the people who were screened and those who were not in terms of prostate cancer mortality.  They found no difference.  There was a 20 percent reduction in mortality observed in the European trial.  So we have two trials with somewhat conflicting results and it's interesting that really what they've done is sort of fueled the debate as opposed to end the debate.  So we continue to have to watch these trials.  We will have follow up data.  People are a little bit concerned about the PLCO in that so many men on the no-screening side went ahead and got screened anyway.  They were not quite sure what we're gonna learn from that going forward but we are watching it very carefully.  Here's a little bit more information about the European trial.  So if you think about the results in terms of how many men do you need to screen to prevent one prostate cancer death?  You can express the results that way, and the numbers look something like this.  So you need to screen about 1400 men and of those 1400 men, about 48 will subsequently be treated for prostate cancer.  And of that 48, 1 man will actually be saved from dying from prostate cancer.  So it's a fairly large number in terms of the number needed to screen.  Now, some people recently have sort of challenged that maybe that number isn't quite that big, but if we go just with the European results, you can kind of get a sense of what the yield is when we screen, versus we treat, versus we prevent death.  So again, it's just fueling the debate is what it's doing.  Well, a colleague of mine at the Mass. General in Boston and also with the Foundation for Informed Medical Decision Making wrote an editorial that accompanied these 2 trials in the New England Journal of Medicine.  And he said some really interesting things I think.  He said the implications of the tradeoffs reflected in these data, like beauty, will be in the eye of the beholder.  Some well-informed clinicians and patients will see the tradeoffs as favorable.  Others will see them as unfavorable.  And as a result, shared decision making or informed decision making for prostate cancer screening seems more appropriate than ever before.  So again, we really haven't ended the debate.  We've just added to the debate. 

Okay, well, one of the questions I was asked in preparing for the talk today is have there been changes in screening rates for prostate cancer as a result of these 2 trials coming out?  Let me be more accurate about that.  Have there been changes in screening rates during the period before the release of the trial compared to the period after the release of the trial?  We don't know if there are results or not, but we do have some data.  There's at least one study that was done among men who were seen in VA, Veterans Affairs Clinics on the West Coast, part of a large organization there.  And what we have are data that are stratified by age--see if I can point to this here.

[ Inaudible Remark ]

>> Yeah, there it goes.  So we have 3 age groups.  They looked at men who are 40 to 54.  They looked at men who were 55 to 74 and then we have men over 75 years.  And you see these 2 time periods here.  So the 2009 group roughly corresponds to the group that was screened before those 2 screening trials were released and then the 2010 group corresponds to me who were screened after those trial results were released.  And it's interesting if you look over at the right side, the text that's in yellow, they have observed some reductions in prostate cancer screening rates.  So these are men who are screen eligible coming in and they've seen a slight reduction in screening rates and as across all those different age groups.  So it isn't just the older men, it isn't just the young men, but across all those age groups.  So it sort of begs the question, what's going on there?  Are the screenings not being offered?  Is it that men are making decisions to not be screened as a result of these new results?  We don't know the answer to that, but when we look at the current trends, they do suggest a slight downward--a downward turn. 

Okay, so updating the prostate cancer screening guidelines, so Dr. Davis mentioned I was part of the American Cancer Society and Prostate Cancer Group that looked at the their 2001 guideline in light of these new results from the 2 large screening trials and one thing that we talked about is the 2001 guideline was often misinterpreted when it was picked up by the media.  It was misinterpreted as saying that the American Cancer Society endorses prostate cancer screening.  And part of that we think is because of how the guideline was initially written and sort of that first sentence in the guideline, the PSA testing and DRE should be offered annually beginning at age 50.  There's something about that initial statement of the guideline that would get picked up by the media and they wouldn't tell the rest of the story and the rest of the story was that there should have been a discussion taking place between the healthcare provider and the patient about the potential benefits, the limitations and the harms associated with testing.  So it's a degree of emphasis that we were concerned about in looking at the old guideline in light of the new evidence. 

So, in updating the 2001 guideline, a number of things happened.  We placed a greater emphasis on informed or shared decision making.  There was--the term “offered” was eliminated because there was concern that that was confusing to people.  There was detailed information added on what should be discussed and the new guideline promotes the use of patient decision aids and I'll talk about those in just a minute, but those are tools to help patients learn about harms and benefits and be prepared for a discussion with the healthcare provider about whether or not to be tested.  There is guidance for men who are undecided.  There's guidance for community-based screening programs and then there's specific recommendations for providers who choose to--for providers whose patients choose to be screened in terms of what you should talk about and then what test you should offer and how often. 

So, the guideline was published in the journal Cancer Journal for Clinicians and here is the guideline and I'm just gonna read the first sentence here.  Again, we spent a lot of time with this.  We wanted to get that first sentence right 'cause we knew that's what was gonna be picked up by the media.  So the American Cancer Society recommends that asymptomatic men, if you recall Dr. Beaver's discussion earlier about screening is related to--is men who are asymptomatic.  Men who have at least a 10- year life expectancy should have an opportunity to make an informed decision with their healthcare provider about whether to be screened for prostate cancer.

>> After receiving information about uncertainties, the risks and potential benefits associated with prostate cancer screening, and then we added another sentence to affirm that prostate cancer screening should not occur without an informed decision-making process.  We spent a lot of time with that to get this message this across that we wanted men to have an opportunity to make a decision.  There's pretty good evidence that at least historically, there wasn't a discussion going on.  So we made that very clear upfront. 

Okay, there's also guidance offered about asymptomatic men with less than a 10-year life expectancy should not be offered screening and also there's some guidance about community-based programs.  The reaction to the guideline actually was very favorable.  The new services that picked it up, we thought they got it right.  In fact, Dr. Davis was interviewed by US News and World Report about this and then also I mentioned Michael Berry from the Foundation for Informed Medical Decision Making.  So we felt pretty good that we got the message right and I will mention that this is very much in contrast with the experience of the US Preventive Services Task Force when they released their guideline about mammography screening for women in their 40's.  They had a very different approach to things and had a very different reaction to it as well.  Even though if you read that guideline, informed decision making is in there, but it's kind of buried. 

Okay, so let's ask a question now, you know, prior to these trial results and updates from the guidelines, were patients already making informed decisions about prostate cancer screening?  We have a little bit of data about that.  This is a study I did with physicians who are members of the American Academy of Family Physicians National Research Network.  So this is a group of family doctors who've agreed to be part of a research network and this is a national network of docs, and we just asked them, how do you handle informed decision making for prostate cancer screening in your practice?  And the docs seem to sort out into these different groups.  The largest group of physicians said that they do indeed discuss harms and benefits of prostate cancer screening with their patients and they let the patient decide and that was about 50 percent of the participants in the study.  There was another group that would discuss harms and benefits with their patients, but ultimately they would recommend for the test.  So when they were asked by their patients, "Doc, what do you recommend?"  The physicians would recommend that they be screened.  But those discussions were going on.  They captured about another 20 percent or so.  There is about 25 percent that said they would screen without a discussion so they don't--they were saying that they don't talk about potential benefits and harms.  They just go ahead and screen their patients.  So if we think about the informed decision making standard, we may wonder if that's indeed happening in those situations.  And then there's a group about 5 percent that don't discuss or recommend screening at all or discuss and then recommend against screening.  My take on that is--and that's about 5 percent of the respondents, those are physicians that are following the old US Preventive Services Task Force guideline and they just are not keeping up with the current recommendations about what should be going on.  So that was about 5 percent of the docs. 

Well, what do the patients say? There was a large national survey done back in about 2007 that's called the decision study.  And this was a random sample of men--actually, men and women across the US, but they pulled out 375 men from the survey who said that they had undergone PSA testing or had a discussion about PSA testing with their doctor in the past 2 years.  And then they were asked questions about, "Well, what were those discussions like?  What did you talk about?"  Seventy percent of those respondents said that they did indeed have a discussion with their doctor about prostate cancer screening, so about 7 in 10 of those 375.  Also about 7 in 10 said that they discussed with their doctor the advantages of screening.  So the advantages were brought up, you know, the reasons to be screened, but only about 1 in 3 said they actually had a discussion about the cons, or the downsides.  So while they're saying that the discussions are occurring, those discussions seem to be largely around the advantages of being tested and not the potential harms of testing.  So the discussions probably are not quite meeting then a threshold for what we would call informed decision making where you talk about both benefits and harms.

Okay, so let's talk about patient decision aids.  So recall the American Cancer Society has recommended the use of patient decision aids to help patients prepare for a discussion about prostate cancer screening with their doctor.  So prostate cancer screening is part of a broader class of decisions that we call preference-sensitive decisions.  These are decisions where the patient's preferences really drive what should be done.  Now we can actually demonstrate that with some of the mathematical models that we build where we build in all of the different treatment options, all the different screening options, how effective they are and then we account for how patients feel about the outcomes related to those and we can show that what's driving the model are those preferences.  And in the case of prostate cancer screening, men feel differently about it.  So this is a preference-sensitive decision.  There are no--there are clear tradeoffs between the harms and benefits and that patient's preferences are central for driving the decision. 

So what are patient decision support technologies or patient decision aids, so what are these?  These are interventions that help people think about the choices that they face, they are interventions that provide information about the options and often include discussion about the option of taking no action.  These interventions help people deliberate about the options.  They get them ready for a conversation with their healthcare provider and they help people forecast how they might feel about short and long-term outcomes, so they give them a sense of what it will be like if they go down one path versus another for them to help them anticipate what those outcomes will be like.  And then the term constructing preferences is one that we've been offering where people sort of learn about the information and start thinking about, "Well, is this important to me.  Does it apply to me or not?  And if it does, how might it impact what I wanna do." 

They are not designed to promote one option or another and they are not meant to replace the physician consultation and that's a really important point because these tools now are being disseminated on a very broad scale and there's some concern that we have to ensure that there is indeed the opportunity for deliberation with the physician in making the decision, that the tool is not in and of itself enough.  They are not meant to stand alone. 

So the basic framework with the decision support is we inform, help the patient deliberate, help the patient decide.  Now, there has been a nice review of cancer-related patient decision aids that include not only prostate cancer but other cancers as well, both screening questions and treatment questions.  And the bottom line from this review showed that these tools do increase patient's knowledge.  They do learn core facts about the options that they're facing.  The tools do not raise anxiety.  There's been some concern that by educating people about all their options, they're going to become so anxious that they actually can't make a decision, that there's not evidence to support, but it is a concern and has been looked at.  The tools also reduce decisional conflict.  In other words, they have people become more assured about what they want to do, so we have good evidence for that.  There's a large review that was done recently of all decision aids, not just cancer and the number of randomized trials is approaching 100 and the results are really very consistent that we have great evidence that patient decision aids are helpful.  There's also language in the Healthcare Reform Act that talks about the use of patient decision aids for these very preference-sensitive questions, it needs to be part of the healthcare experience and a reimbursable activity for clinicians. 

So there are a number of decision aids out here.  I wanted to show you at least one example today and I'm gonna show you the American Cancer Society decision aid.  It's a very good one.  The Centers for Disease Control has a downloadable pdf booklet that you can get.  They have a tool that's specific for African-American man.  They also have a Spanish language tool that can be used.  Mayo Clinic has wonderful tools. There's one other, the University of Cardiff in the UK and we certainly are talking about having a tool developed and available at the Cancer Prevention Center here at MD Anderson as well. 

So, these are some screenshots from the American Cancer Society tool and I was one of the advisors on developing this tool.  It's a downloadable pdf booklet.  You can order them from me or see as you can also go directly to their site.  It's also available online if you want to view it that way. 
A couple of things we did with this tool and I apologize for the orange.  It wasn't my choice.  Being from Texas, I don't think they appreciated why that was important [laughter] but--so anyway.  They liked orange.  Yeah.  So messaging, we had very simple messaging.  We highlighted different things that we wanted to communicate.  For example, you see a statement here in the box, should I be tested for prostate cancer?

>> There may be benefits and risks of the testing.  Research has not yet proven the benefits that testing outweigh the risks, so again reflecting back to those 2 screening trials we saw earlier. 

We also spent some time communicating probabilistic information and what we have here are icon arrays which we have found through risk communication research are a very effective way to communicate.  Probabilities are how often certain things happen.  So we have an icon array showing the number of men who will develop prostate cancer in their lifetime beginning at age 50 and that's the middle--I keep pointing at it.  I'll use my pointer here.  That's the 3 in a 100 in this array here.  That's an incidents rate--the--I'm sorry, that's the mortality rate that's 3 in a 100.  The incidents rate over here is 17 in a 100.  So the principle behind that we're trying to communicate as many more people--many more men will die with prorate cancer than of prostate cancer but it's also a common condition and has a significant mortality rate associated with that.
This is an icon array trying to help us--trying to help communicate the issue of interpreting your actual PSA level.  So if you have a level before--below 4 or above 4, how likely is it that you actually do have prostate cancer? 

We also some guidance here about advantages and disadvantages just in 2 different lists.  The possible benefits to you, for example, testing may find a cancer early while it is small and before the spread.  If found early, there is a better chance of being treated and cured.  Getting tested may give you piece of mind so we go ahead and list those things out so people understand those advantage and those disadvantages. 

And then we have some deliberative guidance also in the tool about questions to talk to others including your doctor knowing the facts and so on.  And then we mentioned this balanced scale which is here.  So this is a tool at the very end of the decision aid where a man can go through and sort of check things that are important to him.  The reasons to be screened or the reasons not to be screened, also indicates some other reasons that are important to him. 

There's a leaning scale at the very end which is a nice tool to sort of give a sense of where you're leaning right now.  This is meant to be taken in to a visit with the clinician to help sort of spur the discussion about whether or not the man wants to be screened. 

Okay so, do patient decision aids impact prostate cancer screening rates?  That question comes up all the time.  So we know that they help men make better decisions in terms of them being informed, more assured, less anxious, do they impact the actual screening rates as well? 
So, this is a system--results of a systematic review I did that was published in 2007 but we only went and looked at studies up to 2006, so I don't have data on studies since, for example, the trials came out which is going to be very important to look at because we have new information that needs to be included in the aids.  But like decision aids in general, we found that prostate cancer screening decision aids increase knowledge.  They increase patient's involvement in the decision.  They become more active partners.  They decrease decisional conflict.  The evidence about the impact of the aids on actual screening rates is somewhat mixed. 

Four men who are just presenting for routine care, coming in for say an annual physical, it does appear that fewer men are opting for screening.  But at the same time, we have evidence that they're making an informed decision about that, but there does appear to be a slight reduction in interest in screening among unselected patients.  That's in contrast to men who are coming in for example for free PSA screening services, or presenting to the doctor for a screening.  The aids have no impact on screening rates among those men.  But for men who are just presenting for routine care, there seems to be a reduction. 

So some concluding comments, and then I'll ask for questions.  So I mentioned professional organizations are generally in agreement about informed decision making for prostate cancer screening.  It talks specifically about American Cancer Society, but that is also in harmony with US Preventive Services Task Force as well.  Decision aids can be helpful in making informed decisions and we really don't know yet the impact of the aids on the actual decisions, but they do appear to promote informed choices and that men are understanding harms and benefits in making decisions that are consistent with our personal values.  So why don't I stop there and take questions.

[ Applause ]

>> So for this segment, because of the recording, we'll need to do questions at the mic if you don't mind getting up.  We have 2 wireless mics in the middle, otherwise, I can repeat them if you feel like you can't make it there.

[ Laughter ]

>> There were some presubmitted questions that relate to this.
>> Okay.

>> And Theresa, if you want to come up I’ll have you address a couple of them.  Really the one that overlaps with both of you--well, this is more technical and the question is “at what age do the men have their prostate exam or screening?”  You saw from Dr. Beaver's, well, how would you answer that?

>> So--and I think it was also laid out in Dr. Volk's.  We typically recommend after informed and shared decision making discussion, average risk man begins screening at age 50.  If you're an increased risk, African-American or family history would begin at age 45.  If there's even greater risk, we may look at earlier, like age 40.  In other words, multiple family members with prostate cancer might look earlier.

>> The next related question was probably really for me I guess.  “What are the signs and symptoms of prostate cancer?”  You know, Dr. Beavers mentioned that in her clinic, they're mainly looking at people without signs and symptoms.  If they get to a urologist--actually, the way I would answer that is the most common signs and symptoms that get a man to a urologist are lower urinary tract symptoms, meaning slow urination, difficulty urination, getting up at night, that kind of thing.  Those are actually usually not prostate cancer related.  That's usually benign enlargement but in the process of checking that out, we may discover prostate cancer so,  I don't know if you have a term for that but I just call that, you know, indirect screening, if you will.  It was screening that occurred because of looking at something else.  It happens all the time actually.  Not the finding the cancer, just, you know, the indirect part. 

The--other one I thought was interesting.  I like this one, really for Theresa as well.  “Is there a general test for any type of cancer, or do you have to have specific test for specific cancers?” is the way they worded it.

>> Yeah.
>>A sort of capture all.

>> There's not one test covers all.  You pretty much have to have targeted tests to look for different diseases.  Some of them are biomarker or blood test based like the PSA.  Some are clinical exams like the digital rectal exam, or a breast exam for a woman and some are imaging tests like the mammogram or virtual colonoscopy, or CT scans.

>> Okay.  So we have some mic questions.  Lets' do those.  Sir, you're first.

>> Yes, question, I saw a lot of statistics about the US numbers, you have any that compare that with other countries like France for instance?

>> Sure.  Well, there're incidents and mortality statistics and they--for the most part, industrialized nations have the similar ratio on their populations.  The big exceptions that gets everyone's attention would be if you look at Asian populations.  They eat, you know, a lot of rice, fish, low-fat diets so they tend to have lower incidents of cancer screening.  A little of that is still confounded by the fact that many of those countries don't screen as aggressively as we do.

>> That was my point, correct.

>> So overall, cancer detection rates in China for example are very low and that could be a mixture of looking forward as well as their diet.  The classic epidemiologic teaching is that if Asians migrate to the US and assume our dietary habits, then their rate comes up to the risk of everybody else, so.

>> One of the risk factors that showed up on that one slide said that even though your PSA test is low, you might still have prostate cancer.  Is there any data that kinda indicate how great a risk that really is?

>> At the most, and I think we've shown the slide at prior meetings so you could look at them on the archive websites, but the test that got our attention was a screening trial where men got put on a preventative agent versus a, you know, placebo or dummy pill.  And then at the end of the trial, everybody got a biopsy.  And it was actually one of the first trials where we had several thousand men with what we considered low PSA's get a biopsy anyway.  And that's why a lot of us have lowered the PSA threshold.  So even for PSA between 2 and 3, you're seeing cancer detection rates in the 20 percentile.  And really even if the PSA was 1 to 2 or less than 1, you still had detection rate.  So there was no normal PSA and any group of men that got diagnosed, you know, a small percentage of them had high grade cancer even.  So it's a range of risk and that's part of, you know, the education process.

>> Thank you.

>> For effective prostate screening, how important is it to have the PSA and the digital rectal exam or can you just have the PSA?

>> Well, that's kind of related to the prior answer in that some men have a low PSA and still have an abnormal exam.  The PSA is higher yield without a doubt, but some tumors are discovered by the exam alone.

>> Okay.

>> So, it's better to do both, but if you had to pick one, it would be PSA, so.

[ Laughter ]

>>Fair enough, right?
>> Yeah.

>> Good morning, sir.  I wanted to find out from you what are the factors that contribute to higher PSA apart from the biological factors?  Do social factors contribute to higher PSA?

>> Good question.  Do you have a list that you gave in the screen for me?

>> Of social factors?

>> Well no, everything but cancer that rises the PSA.

>> No, I mean we don't have one in the clinic but--

>> It's a huge list and but it's anything that abnormal with the prostate, benign enlargement, infection, you know, people looked at bicycle riding, any sort of trauma.  Recent intercourse, they bump it for a little bit.  But you know, the biggest ones are the cancer and the benign enlargement.  So there are some men that just have high PSA's and they don't have cancer but sometimes we have to look anyway, so. 
I know there're a lot of men standing, we do have one more session.  Why don't we do one more question and then let's do the cases and we may get to some of your questions and if that's okay.  Let's do one more?

>> Okay.

>> And just try to distribute it well.

>> Hi, I hope this is part and parcel, but could you have something to say about proton therapy?

>> Yeah, well we have an expert in the next session.

>> Thank you.

>> Good timing.  You're in the right place, so.  I'll just last--Dr. Volk to make maybe one more comment, what's the agenda for the next, you know, few years for patient education now that we have the trials, you have these tools.  What's on your radar screen for what you want to accomplish with this in the next few years?

>> Right.  So the real challenge I see facing us next is actually implementing these tools in routine clinical care.  So we know how to develop the tools, we have the sense of how they work, we're not very good at getting them used in the clinical setting so we're working on that now.  You know, many practices for example have now moved to computerized patient records and often they'll have computers in the exam rooms.  That's one opportunity for us to actually disseminate some of the educational materials that way, is just have these tools available sort of right that point of care where people can experience them.  So that's what where we're working now is trying to get this implemented on a much broader scale.

>> And then for Theresa, I guess you mentioned getting all the guidelines out on the web and as well as some new guidelines for lung and other sites, so.

>> Yeah, yeah.  We're gonna have 9 different guidelines.  We currently have breast, cervical and colorectal.  Coming down the pike are lung, endometrial, ovarian, liver, prostate and skin.  So we'll have very specific recommendations for average and increased risk and I hope you'll go the website to understand your cancer risk and what screening tests are appropriate for you.

>> Good, alright.  Well, let me thank our 2 featured speakers and your questions and we'll do our changeovers.

[ Applause ]

>> Thanks very much, good.  So--