This forum is for the discussion of anything to do with Prostate Cancer. There are only four rules:
No fundraisers, no commercials (although it is OK to recommend choices of treatment or medical people based on your personal research; invitations to participate in third-party surveys are also acceptable, provided there is no compensation to YANA);
No harvesting e-mail addresses for Spam;
No insults or flaming - be polite and respectful at all times and understand that there may be a variety of points of view, all of which may have some validity;
Opinions are OK, but please provide as much factual evidence as possible for any assertions that you are making
Failure to abide by these simple rules will result in the immediate and permanent suspension of your posting privileges.
Since this is an International Forum, please specify your location in your post.
The Problem with PCa Mortality Statistics (My Own Personal Perspective)
First of all, this is not intended to be a continuation of the friendly disagreement with Terry on the issue of age & AS from last week. That is over & done with. Plus, I can’t disagree with Terry when he said: “Young men have to live long enough to die from PCa - and many will not make it” because, by definition, the word “many” means a large, but indefinite number. So, the number is subjective & open to interpretation. Also, like Terry, I have read that, overall, less than 3% of male deaths are from prostate cancer. In addition, in further researching this subject, I found that, according to The American Cancer Society (ACS), a man has about a 1 in 36 chance of dying of PCa in his lifetime in the USA. When you compare these two statistics, the 3% figure appears to be reassuringly low, while the ACS figure seems rather high, but in reality 3% & 1 in 36 are basically the same (something my wife needed to point out to me).
That is one of the reasons why I don’t care for statistical analysis. Personally, it has always been hard for me to grasp. (I even flunked the course in school despite being a good student). Today, though, I dislike it for another reason. When used in the discussion of PCa mortality, I find statistics on the subject to be cold & impersonal. For example, as I said, the overall 3% PCa mortality rate listed above sounds encouraging, but that 3% is made up of thousands of individuals whose lives were cut short by the disease. For example, in 2011 the ACS estimates that about 33,720 men will die of PCa. It bothers me when I see the lives of these men expressed in terms of statistics designed to show that having PCa is not that great of a risk.
So, when it comes down to assessing the risk of dying from PCa, I think that, rather than expressing that risk in terms of a ratio or a percentage of the population, it should be assessed by each individual patient, looking at his circumstances alone. For example, if a 48 year old man, like Andy C., needs to decide if treatment is right for him now or in the future, IMO he would be far better off looking at his age, overall health & biopsy results rather than relying on seemingly low PCa mortality rates. In other words, instead of looking on the 3% mortality figure as a positive, it would be wiser of him to objectively look at his own circumstances and ask himself if he is just as likely to be one of the 3% who die of PCa as one of the 97% who die of something else. For example, if his circumstances are such that he is in good health (aside from having PCa), then at his age he can reasonably assume that he has almost 30 years of life expectancy left, and with that knowledge, he should take a good hard look at his biopsy results. If the results show a higher risk cancer, like Andy’s, then he COULD very well end up being one of the 3% if he chooses to forego treatment in favor of an AS strategy. This is because, as we all know, a higher risk cancer has a greater chance of spreading and, if it does, then he is highly unlikely to live out the almost 30 years of life expectancy that he would otherwise have left. The bottom line then is that PCa does not fit into an easily defined mold as far as risk is concerned and it should not be viewed that way.
Lastly, in a similar light, it would also be unwise to put much stock in the fact that “many” people aren’t going to live to the age of 76 anyway, (which is the average life expectancy for a male in the USA). I say this because, if your overall health is good, then odds are that you are not going to be among the many who are going to die prematurely & you should therefore act accordingly.
Re: The Problem with PCa Mortality Statistics (My Own Personal Perspective)
Alan, I agree with a lot of what you say. In talking to my oncologist today, I was reminded that the stats are made up with a high proportion of older men (ie in their late 70s and beyond) having multiple problems, perhaps a history of smoking for many years, often unfit and (especially in the US but increasingly so, alas, here in the UK) overweight. Those of us who are fit and well apart from the PCa need to bear in mind that we are better equipped than most to handle the op or EBRT and/or chemotherapy and so beat the stats and outlive our PCa.
Put another way, keep fit, eat healthily and think positively and you'll probably do better than most!
Re: The Problem with PCa Mortality Statistics (My Own Personal Perspective)
Alan,
Your post has prompted me to get down to what I have been meaning to do for some time now - write a piece on why I believe that age is not the key factor in making a decision regarding treatment or non-treatment.
I use available statistics because we have to make our decisions within the context of the overall statistical 'odds'. I was statistically inept when I was diagnosed (I hadn't flunked a statistics course because I never did one) but I found a piece that started me on the track to gaining a better understanding. The piece is THE MEDIAN ISN'T THE MESSAGE by Stephen Jay Gould and I hope you find it as useful as I did.
Gould's message is clear - the median or average number in any set of numbers (life expectancy in this post) has a function, but what is more important is the range of the numbers and where we are likely to fit into the range.
Essentially each of us has a life expectancy which extends from about ten minutes from now (roughly how long it takes you to die from a heart attack) to age 120 - because no male has been recorded as living longer. Somewhere between those two is the average length of time a male in the USA is expected to live. That length of time changes as a man ages. Clearly, if the average life expectancy is 76, then that age cannot apply to a man of 77.
Using the tables suggested by NCCN the average life expectancy of a man of 45 in the US is 33 years, taking him out to age 78. Although I used the word 'many' in my previous mail, perhaps I should have been clearer. Because the age of 78 is the median or average, this means that half the men of 45 in the USA today will die before the age of 78 if this table is correct. The other half will live on, some for many years. The life expectancy of the man who makes 78 is about 10 years, taking him out to about 88. There is even a calculated life expectancy for a man aged 100 - two years.
The NCCN also suggest that it is reasonable to take into account the overall health of the man when making any decision which involves trying to estimate life expectancy and that the normal 'range' should be 50% each side of the median or average. So that means for the man of 45, they are suggesting that his range of life expectancy is between 16 and 49 years - ages 61 to 94.
Those are the broad statistical issues I found it useful to understand and to review. I then started trying to work my way through the PCa statistics to gain an understanding of how my specific diagnosis might affect my lfe expectancy. The answer I came up with IN MY SPECIFIC CASE was - very little.
I don't have time to show how the overall life expectancy figures fit in with the PCa figures right now, but will cover that in the piece which I will try to write before I got off to South Africa to celebrate the fact that I have beaten my life expectancy already:-) Suffice it to say that 95% of deaths from PCa occur in men over that age of 60. This means that 5% of the prostate cancer deaths which account for less than 3% of male deaths in the US are statistically likely to occur in a man of diagnosed at age 45 in the following fifteen years. Put another way, if we look at 1000 US male deaths, less than thirty men will die from prostate cancer and of those thirty men only one or two will be under the age of 60 – and about 18 of the deaths will be over the age of 80.
There are clear pointers as to which unfortunate men may constitute the very small number of men who die under the age of 60. Men with a high Gleason Score of 9 or 10 are at the greatest risk. The AS studies which are running now show that for men with a similar diagnosis as the men in the studies the risk of disease progression are very small and the risk of death even smaller.
One last thing - and then I must go. When I found that the best life expectancy anyone would give me was about 15 years (more likely 5 years) I looked back fifteen years and thought "Could I have predicted what we would do in the fifteen years that followed? Can I predict what will happen in the next fifteen years?" I realised I was trying to hard to predict the unpredictable and went back to the broad statistics and tried to figure out where I fitted in.
I hope this helps a little. Oh! One last thing - it is awful that about 30,000 men will die in the next 12 months from PCa, but more than one million men will die of other causes and more than half the men who die from our shared disease will, as I indicated above be over the age of 80.
Re: The Problem with PCa Mortality Statistics (My Own Personal Perspective)
I will try this example to help you understand the statistics.
You are in a class with 36 other boys. Thirty decide to play on the field and six of you decide to play on the road including yourself. Knowing that one of the boys will get killed by a car what are your chances of getting killed by the car?
Ignoring all other factors, once you are diagnosed with PC you are one of the boys playing on the road.
Other factors will change the odds for each boy on the road. If you have good vision and can dodge cars better than the other boys then you can improve your odds of surviving.
Re: The Problem with PCa Mortality Statistics (My Own Personal Perspective)
Interesting analogy, Frank, but it ignores some factors, including any idea of what game the 30 boys are playing on the field. If for example they are playing Marines at War with their parents' shotguns, they might be in more danger collectively than the boys playing kick ball in the quiet one way street outside the school with the crossing guards looking on:-)
Statistically men diagnosed with prostate cancer are as likely to die from the causes of death that the non-diagnosed men will die of. Of course this doesn't apply to all individual cases and it is not possible to predict with accuracy which of the PCa men will die from the disease over the next 30 years - even the men who choose early therapy are never at a nil risk.
Re: The Problem with PCa Mortality Statistics (My Own Personal Perspective)
My example was only to illustrate the chances of dying from PC once you get diagnosed with it. We all die eventually but the boys on the field will never die from getting hit by a car (getting PC).
Once you are diagnosed you are now standing in a group of six men, one of which will die from PC and not the original group of 36.
I am actually a big active surveillance fan for GS 6 patients but it is good to understand the statistics.
Re: The Problem with PCa Mortality Statistics (My Own Personal Perspective)
I'm not too sure that I've seen that 16% mortality rate, but assuming it is correct and in the light of my glass half full look at life then it must be said that of the six men, five will not die of prostate cancer, but they will die of something else - and maybe before the man with cancer.
'Curing' prostate cancer does not make anyone immortal.
Re: The Problem with PCa Mortality Statistics (My Own Personal Perspective)
Warwick,
I found it extraordinarily difficult to find good data on mortality when I was diagnosed. I wanted to get some idea how long I might have, how would I die, how soon…..round and round in my head. I couldn’t find anything, no one would discuss the issue. You can try yourself on one of the many prostate cancer forums and you’re likely to find no responses. That is why I wrote The Elephant In The Room . I don’t claim that everything is statistically correct in that piece, but, I feel, it is better than the deathly silence that greets most enquiries on the subject and enables people to gain some insight into the issues. As you will see there are two studies which do not support the 1:6 mortality rate for all men with prostate cancer.
But to your specific query about the variations in survival related to Gleason Grades, I can say that again there are very few long term studies that are of much use to men diagnosed today. The Abstract of the one that I found fairly early is set out below, but can I ask anyone who reads that to understand some key issues:
1. The study was published in 1998 and although some of us feel that was yesterday, in fact it is over thirteen years ago.
2. The men in the study were “young” since they were in what was then the lower half of the range of men sorted by age – the median age for diagnosis being about 74 before the introduction of PSA testing
3. The men were diagnosed before the introduction of PSA testing. They would have had biopsy procedures because they had symptoms, a positive DRE or from material recovered in a DRE (Digital Rectal Examination). In that era about 80 – 90% of men had advanced disease when diagnosed. Currently about the same percentages are diagnosed with early stage disease.
4. As will be seen, the old Gleason Grades were still in use. Based on Gleason Grades on a scale of 1 thru 5, the range of Gleason Scores was from 2 thru 10, with GS 6 as a median. Gleason Scores below 6 from needle biopsy procedures are no longer labelled “prostate cancer”. GS 6 is the entry level now. This practice has led to what is termed “Gleason Migration” . This means essentially Gleason Scores have moved to higher levels as demonstrated by specialist pathologists grading old specimens and comparing current grades against the initial ones. Scoring and grading pathology specimens is a subjective process.
I’d be very interested to see any other studies that provide better information from a later era, but fear that this is not available .
Here’s the study abstract:
Albertsen PC, Hanley JA, Gleason DF, Barry MJ.
Competing risk analysis of men aged 55 to 74 years at diagnosis managed conservatively for clinically localized prostate cancer.
JAMA 1998 Sep 16;280(11):975-80
CONTEXT: The appropriate therapy for men with localized prostate cancer is uncertain. Until results of clinical trials are available, men and their physicians need guidance.
OBJECTIVE: To estimate survival based on a competing risk analysis stratified by age at diagnosis and histologic findings for men diagnosed as having clinically localized prostate cancer and who were managed conservatively.
DESIGN: Retrospective cohort study.
SETTING: Connecticut Tumor Registry.
PATIENTS: A total of 767 men with localized prostate cancer diagnosed between 1971 and 1984, aged 55 to 74 years at diagnosis, either not treated or treated with immediate or delayed hormonal therapy, and followed up for 10 to 20 years after diagnosis.
MAIN OUTCOME MEASURES: Estimates of the probability of dying from prostate cancer or other competing hazards.
RESULTS TABULATED:
• Men with tumors that have Gleason scores of 2 to 4 face a 4% to 7% chance of dying from prostate cancer within 15 years of diagnosis depending on their age at diagnosis.
• Men with tumors that have a Gleason score of 5 face a 6% to 11% chance of dying from prostate cancer within 15 years of diagnosis depending on their age at diagnosis
• Men with tumors that have a Gleason score of 6 face a 18% to 30% chance of dying from prostate cancer within 15 years of diagnosis depending on their age at diagnosis
• Men with tumors that have a Gleason scores of 7 face a 42% to 70% chance of dying from prostate cancer within 15 years of diagnosis depending on their age at diagnosis and
• Men with tumors that have Gleason scores of 8 to 10 face a 60% to 87% chance of dying from prostate cancer within 15 years of diagnosis depending on their age at diagnosis.
CONCLUSIONS: Men whose prostate biopsy specimens show Gleason score 2 to 4 disease face a minimal risk of death from prostate cancer within 15 years of diagnosis. Conversely, men whose biopsy specimens show Gleason score 7 to 10 disease face a high risk of death from prostate cancer when treated conservatively, even when cancer is diagnosed as late as age 74 years. Men with Gleason score 5 or 6 tumors face a modest risk of death from prostate cancer that increases slowly over at least 15 years of follow-up.
PMID: 9749479 [PubMed - indexed for MEDLINE]
Anyone confused by the statistical jungle that is part of the prostate cancer world mught find this study intriguing Non-prostate-cancer mortality risks among men with prostate cancer since it suggests that men who are diagnosed with clinically detectable prostate cancer — but do not actually die from their prostate cancer — are at higher risk for all other cause mortality than men who show no signs or symptoms of prostate cancer.
Good luck on working that one out, although I must say that one of the early published analyses of the well known long term Swedish study (which looked at delayed versus immediate treatment) also showed that in the first ten years there was no statistical difference between the overall mortality rates of men in either arm. Slightly more men in the 'untreated' arm died of PCa than those in the 'treated' arm, but this was balanced by a greater number of non-PCa deaths in the 'treated' arm.
Over 20 years, this difference grew wider and produced an indication of advantage for treatment over non-treatment, although PCa deaths in each arm were a minority cause of death. Interestingly enough, there were still biochemical failures being recorded in treated men right up to the end of the study at 20 years out.
Terry, this statement: “Interestingly enough, there were still biochemical failures being recorded in treated men right up to the end of the study at 20 years out.” does not come as a surprise. In a long-term study of 3500 men who underwent RP at Johns Hopkins, it was found that 71% had an undetectable PSA at 20 years. (The study concluded that these figures would be even better today since 500 of these men were from the pre-PSA era).
These figures were reported by Johns Hopkins to show the POSITIVE outcomes of RP in men with organ-confined disease. So, like you, they are looking at these results from a glass half full perspective (or, in their case, it would be more accurate to describe that glass as almost three quarters full after 20 years).
One thing that should be noted is that these figures are likely to be a best-case scenario since Johns Hopkins is one of the best cancer treatment centers in the U.S. One would have to figure that, nationwide, the recurrence rates following RP are higher.
Additional study facts: 1) The probability of maintaining an undetectable PSA at 5, 10 & 15 years was 90, 82 & 78% respectively. 2) Of the men who eventually developed an elevated PSA, it went up in the first 5 years in 58%, between 5 & 10 years in 42%, & beyond 10 years in 10%. 3) In men who were at the lowest risk for recurrence (defined as a PSA <10, stage T1c or T2a, and a Gleason score of 6 or lower), 16% developed an elevated PSA by 5 years, 74% between 5 & 10 years, & 10% after 10 years. 4) Of the 29% experiencing recurrence at 20 years, only 6% involved cancer at the surgical site.
Good question David. The fact is that it adds up to 110%; so there is obviously a mistake. It wasn't mine though because I just re-checked my source & the mistake is there in print. Therefore, I cannnot supply the correct figures, but I don't see it changing the essence of the study results anyway.
Good eye though. You have one-upped this OCD guy. Thanks for the reply.
Re: The Problem with PCa Mortality Statistics (My Own Personal Perspective)
Albertsen PC, Hanley JA, Gleason DF, Barry MJ.
Competing risk analysis of men aged 55 to 74 years at diagnosis managed CONSERVATIVELY for clinically localized prostate cancer.
JAMA 1998 Sep 16;280(11):975-80
I wonder if knowing what is meant by CONSERVATIVELY would help further understanding of this study.
Re: The Problem with PCa Mortality Statistics (My Own Personal Perspective)
"Conservative Managment" meant that there was no immediate aggressive therapy.
The term was probably synonymous with Watchful Waiting at the time and would have evntailed men being treated for symptoms rather than making an attempt at "cure".
It was the path folowed in most countries outside of the USA where the surgeons particularly have always been very aggressive in their treatments. As Dr Christopher Logothetis said in a talk to an US-Too group back in December 1993.
One of the problems with prostate cancer is definition. They label it as a cancer, and they force us all to behave in a way that introduces us to a cascade of events that sends us to very morbid therapy. It's sort of like once that cancer label is put on there we are obligated to behave in a certain way, and its driven by physician beliefs and patient beliefs and frequently they don't have anything to do with reality.
People who come from a totally different social background like the Swedes are very frustrated with us for the frequency with which we do prostatectomies. In fact, they are very angry. And urologists are probably the most embattled surgeons across the country in the United States. If you go overseas, and you are a urology surgeon in the U.S., there is a reputation that you carry with you that you are willing to operate on anybody the minute they sneeze.
Now that's not necessarily true. Actually, the U.S. surgeons are driven by attempts to cure more people. So I think the motives are very correct in the overwhelming majority of them. But one can't ignore the paradox that the earlier you treat, the larger number of patients you do operations on who would do well despite you. So proportionately the patients who benefit, that is those patients who are cured by your operation, start becoming a smaller and smaller fraction of the total. So there's the paradox with prostate cancer.
Nothing much changed from 1993 until fairly recently when there was finally a recognition that Dr Thmoas Stamey was correct when he said in 2001
I believe that when the final chapter of this disease is written, which is unlikely to be in my lifetime, never in the history of oncology will so many men have been so overtreated for one disease.
Re: The Problem with PCa Mortality Statistics (My Own Personal Perspective)
Hello Terry,
It does get more confusing when one reads, as I did, the abstract in JAMA which says the study included men who had no treatment or were treated with immediate or delayed hormonal therapy.
Re: The Problem with PCa Mortality Statistics (My Own Personal Perspective)
Yes, Jack, that is what I said:
"Conservative Managment" meant that there was no immediate aggressive therapy.
The term was probably synonymous with Watchful Waiting at the time and would have evntailed men being treated for symptoms rather than making an attempt at "cure".
Re: The Problem with PCa Mortality Statistics (My Own Personal Perspective)
Terry,
I see where I made a wrong assumption of when this conservative management took place. It would have been clearer if they had said the study included men who HAD HAD either no treatment or immediate or delayed hormonal therapy PRIOR to their being included in the study. One of the limitations of the study mentioned is that some of these patients received aggressive management at a later date. I assume if they had been doing ADT during the period of observation that would not have been considered a limitation of the study. Still, it looks like the findings are valid. So you're saying in 1988 Watchful Waiting included hormonal therapy. Is that still the case? I note that one study showed the outlook for an older man (like me) diagnosed with Gleason 9 was pretty good, but that was later refuted. Darn!
Re: The Problem with PCa Mortality Statistics (My Own Personal Perspective)
Jack,
There is what has always seemed to me to be an artifically created 'divide' between Watchful Waiting WW and Active Surveillance AS.
WW is said, in this type of fine division, to be a choice that mainly treats symptoms and makes no attempt at cure, the aim being to manage the disease with a minimum of loss of quality of life until the man dies - whether of PCa or some other cause
AS on the other hand is said to be a choice that monitors the man as closely as possible to ensure that aggressive action, aimed at curing the man but with the inevitable loss of some quality of life, is taken as soon as there is evidence that this would be in his best interests.
Now those are very broad statemetnts and there are many overlaps and lack of clarity.
So WW would almost inevitably include ADT if the circumstances warranted it and the man lived long enough (me for example!!) as would any conventional treatment that failed to 'cure' the man as a large percentage does at present. Well, if you defined bio-chemical failure as 'failure.'
The reason that older men often show a better disease specific survival rate compared to younger men is that they are more suscetible to death from other causes. You can't die twice.
Re: The Problem with PCa Mortality Statistics (My Own Personal Perspective)
the advantage of being a pca surviour is we may or may not now from what we will die from. If we wake up in the morning we are alive, then get on with the day and enjoy it. Age is unimportant. bob
Re: The Problem with PCa Mortality Statistics (My Own Personal Perspective)
I want to say thank you to David, Terry & Frank for replying to my post.
I appreciate the statistical insight, but I may have given the wrong impression. I do understand statistics in the sense of knowing how to interpret ratios & percentages. An example of what I meant when I said that I found statistics hard to grasp is when I said this in my last post: “When you compare these two statistics, the 3% figure appears to be reassuringly low, while the ACS figure seems rather high, but in reality 3% & 1 in 36 are basically the same”. Similarly, when you look at the ratio supplied by Frank, 1 in 6 seems like bad odds, but when you convert the ratio into a percentage, it is 17% and that seems like a rather small figure (especially when you factor in what Terry said about half of those men being over 80 years old).
That being said, I don’t want my main reason for writing that post to be sidetracked by the responses that, while helpful & appreciated, focus on the perception that I need help understanding statistics. What I really wanted readers to get out of that post is best summed up in the long second to last paragraph. Here is another example of the point I am trying to make: In Terry’s initial response, he said that only half of the estimated 34,000 PCa deaths in the U.S. in 2011 would involve men under 80 years of age. That seems like a very low figure indeed, but I think that it is important that an individual not put too much stock in it. I say this because his individual circumstances MAY result in him being one of the 17,000 fatalities. So, IMO, his focus should be on his, & ONLY HIS, individual statistics & not the national ones.
This is not to say that I think that Terry disagrees with my assessment. I believe that the following statement from his second to last post shows that, basically, we are on the same page: “Statistically men diagnosed with prostate cancer are as likely to die from the causes of death that the non-diagnosed men will die of. Of course this doesn't apply to all individual cases and it is not possible to predict with accuracy which of the PCa men will die from the disease over the next 30 years.”
Lastly, IMO, if somebody was going to give any weight to national PCa mortality statistics, it should be the ratio of 1 death for every 6 PCa patients quoted by Frank. I believe that it is a much more relevant statistic than the often cited < 3% mortality figure that is based on the population as a whole.
