 |
 | Home | Contact Us | FAQ | Search & Site Map | Link to Us |
 | Sign In | Join | Other 45 Sites in Network |
Medical Forum / Diseases and Disorders / Prostate Cancer / July 2007"The proof of the pudding is in the eating."
Some 8 years ago Dr. Charles Myers, a well recognized medical oncologist and expert in prostate cancer treatment was diagnosed with advanced localized prostate cancer that had invaded the lymph nodes. To deal with his aggressive PCa presentation, Dr. Myers decided for aggressive treatment. He chose maximal androgen deprivation neoadjuvant to high dose radiotherapy and a seed implant. Hormonal therapy was continuous for 18 months and then maintained with a 5- alpha reductase inhibitor. Here is a brilliant oncologist, one of the few that totally specializes in the treatment of prostate cancer that like many of us undergoes androgen deprivation as early as possible in an effort to extend his survival. Is it possible that he misunderstands the voluminous medical literature indicating the futility and inefficacy of early or even hormonal suppression at all? I doubt that he would undergo such radical therapies when the evidence is so tilted to support the opposite. That is why I believe that the proof of the pudding is in the eating...The evidence in a survival advantage is there for all to see and it is not just a few months depending on the severity of the disease at diagnosis and on the treatment applied. Do not be fooled by misinterpretations on statistical results. Read Dr. Myers article because it is expressed in coherent layman terms and supports with references what he experienced in his own flesh. http://tinyurl.com/2oqojb Best of health, RalphV www.pcainaz.org/phpbb > Do not be fooled by misinterpretations on statistical results. How de we set about avoidance of such things? > > Do not be fooled by misinterpretations on statistical results. > > How de we set about avoidance of such things? 1. The best way is for you to learn statistics and their significance 2. A second option is to find someone you know and trust, that understand them and explains the results to you. 3. Until you do 1 or 2, doubt the interpretations given here... More so when they claim to have done extensive research. RalphV www.pcainaz.org/phpbb >> > Do not be fooled by misinterpretations on statistical results. >> >> How de we set about avoidance of such things? > >1. The best way is for you to learn statistics and their significance I think grappling with statistical method is superfluous to requirements and in any case would be very hard-earnt. Maybe I'm wrong but I would guess that few, if any, studies have been rejected on a technical nicety that only statisticians would understand. Studies deliver their results in a familiar numerical form so we should only need an idea of relative magnitude to evaluate them. The numbers do become unwieldy though as the number of categories, parameters and curative strategies rise and this latter is a common problem for us here. >2. A second option is to find someone you know and trust, that >understand them and explains the results to you. >3. Until you do 1 or 2, doubt the interpretations given here... More >so when they claim to have done extensive research. I don't know why I should be more doubting the more extensive the research. You seem to have done a lot of research yourself. How should I treat your point of view? Do you not think that in forums like this, through debate, we get as much of a chance at reaching an approximation to the truth as is possible anywhere? >>> > Do not be fooled by misinterpretations on statistical results. >>> [quoted text clipped - 4 lines] > I think grappling with statistical method is superfluous to > requirements and in any case would be very hard-earnt. If you mean that going through all the steps of a calculation is too much to ask of a consumer of data then I'd agree. But I agree with RalphV that it is important to have a precise understanding of what researchers report and what it means. In other threads, I've harped on the idea that just because two variables are associated, that does not mean that they are causally linked. Having ideas like that in mind as you read news reports, press releases, abstracts, and papers on statistical studies would help you better understand what they are saying and not saying. > ... I would guess that few, if any, studies have been rejected > on a technical nicety that only statisticians would understand. And disasters have happened because statisticians were not in the loop to reject things. Remember the space shuttle Challenger? That disaster was completely avoidable! They had the data that would have told them that the o-rings were unsafe under the launch conditions. But there was no statistician consulted. I know some statistical concepts are hard -- I teach the stuff, and I see my students struggle with it all the time. But it's important to know, especially if you're trying to make decisions from data. --charlie  Signature6/2006 PSA 5.2 DRE suspicious 7/2006 Biopsy 2 of 10 positive Gleason 7(3+4) 11/2006 LRP Clear margins 1/2007 PSA < 0.01 3/2007 PSA < 0.01 6/2007 PSA < 0.01 so far, so good >>>> > Do not be fooled by misinterpretations on statistical results. >>>> [quoted text clipped - 9 lines] >is important to have a precise understanding of what researchers report and >what it means. But precision is not on offer in any of the results we see. I'm not sure what you mean by understanding. None of the data emerging from PCa studies should require any specialist understanding beyond competent numeracy and the knowledge that the lower the p-value, the more statistically significant the result. >In other threads, I've harped on the idea that just because two variables >are associated, that does not mean that they are causally linked. Having >ideas like that in mind as you read news reports, press releases, >abstracts, and papers on statistical studies would help you better >understand what they are saying and not saying. I'll admit to being lapsed and rusty but as a maths B.Sc. with a year of post-grad followed by a short spell of teaching maths and physics, I could have been a professional (" instead of just a bum - which is what I am ") had I not been lured elsewhere. No, I would hotly contest the benefit of studying stats to get to grips with what *we* need to know. Your advice, above, against jumping to conclusions is perfectly sound and shouldn't require any elaboration. >> ... I would guess that few, if any, studies have been rejected >> on a technical nicety that only statisticians would understand. (my fault, I should have made myself more clear - I meant to say that to the best of my knowledge, no statistical results relating to *PCa* have ever been challenged on their methodology or statistical validity yet we've no shortage of contentious bones to gnaw on in this forum. I've no doubt plenty of competent mathematicians have been struck down by this disease yet in spite of their self-interest none of them has come forward with a definitive resolution to our argument/s.) >And disasters have happened because statisticians were not in the loop to >reject things. Remember the space shuttle Challenger? That disaster was >completely avoidable! They had the data that would have told them that the >o-rings were unsafe under the launch conditions. But there was no >statistician consulted. No doubt, but we lack no such insight here - we in PCa-world specifically are deluged by stats yet still we founder. >I know some statistical concepts are hard -- I teach the stuff, and I see my >students struggle with it all the time. But it's important to know, >especially if you're trying to make decisions from data. I'm trying to think of anything I learnt or taught about stats/probability/chance that would come in useful in this forum and in particular in this thread. Other than common-sensical items nothing springs to mind. Which elements of your stats course do you think would be of benefit? >...Which elements of your stats course do you think would be of benefit? I wrote what I thought was a very nice reply to this, only to have my computer swallow it up. Arrgh! I'll try again. I teach the standard general-purpose introductory statistics course you'd find at most small liberal art colleges in the US. But I emphasize the theme that consumers of data should be able to think critically about data, and that a general understanding of statistical methods enhances that ability. One of the first exercises that the students do is to read a few news articles reporting on research, and to see to what extent they could answer specific questions based on what they read: who/what were the subjects, who funded the reserach and why, who actually did the work, how were the subjects chosen and measured, what is the magnitude of the result, etc. As we progress through the course, we often return to these and see how we can sharpen those questions. I have no illusions about students remembering specific formulas or the syntax of the computer software we use (Minitab), but I hope that my former students do remember at least some of the following from my course whenever they deal with statistical information: 1. All research has an agenda, some of which may be hidden. It does not render research invalid, but a good reader should be aware of it. 2. Bias is unavoidable, but competent researchers will do everything they can to minimize it. 3. Outliers matter. Never discard them without explaining them first. 4. Association never implies causation. 5. Simulations are useful, but they can never replace actual data. 6. Do not confuse Prob(A, given B) with Prob(B, given A). 7. Creating a confidence interval is like playing ring toss, not darts. 8. A hypothesis test always starts with a yes/no question, continues with one of many statistical processes, and ends with the probability that you are wrong if you answer "yes." You may call that list common sense, but it has been my experience that people who are not at least somewhat familiar with statistics have a very hard time applying ideas like those to information they are given. And in any case, the more statistics one has under the belt, the sharper those ideas can become. As far as this partiicular thread ... I don't know much about ADT, as I have been fortunate not to need it thus far. But I do want to get Walsh's latest edition anyway. Like Steve K., I like the older edition but find it dated. And the studies that others have mentioned are on my reading list now, though I honestly do not know how soon I'll get to them -- I'm still wading through the multivitamin thing I say I'd look over. It seems to me that at least part of the argument here is centered on the outcomes question. If someone wants to keep a tumor as small as possible for as long as possible, and that's more important to them that the potential side effects, then ADT as soon as a rising PSA trend is identified makes sense. If the side effects are not worth the possibility of SEs, then they shouldn't. In any case, it should be the informed patient who decides when the outcome has been successful. I'll have to look over the studies about survival after ADT. Since some are criticizing sampling and measuring of one of the studies and others are saying it was fine, I must see for myself. That's where some stats knowledge will come in handy. In any event, I think we'd all agree that there's a lot that needs to be learned about the biochemistry of the androgen-independent PCa cells. Knowing how to target those buggers is the big prize. --charlie (hoping the computer doesn't eat this one, too!) >1. All research has an agenda, some of which may be hidden. It does >not render research invalid, but a good reader should be aware of [quoted text clipped - 17 lines] >are given. And in any case, the more statistics one has under the >belt, the sharper those ideas can become. Your list certainly includes some common sense cautionary items but I don't see how 6,7 or 8 have any place in our PCa-consciousness other than in the love of study for its own sake (I could make some recommendations of my own!). On the face of it, it looks almost like a deliberate attempt to mystify, since I'm sure you must realise that most folk won't have a clue what your talking about. Given the notoriously approximate and sometimes apparently 'conflicting' data relating to PCa studies, could you demonstrate the relevance of 6,7 and 8 to a patient who needs to do no more than make a rational treatment choice? Once again, my dumb office computer has seemed to eat my original response. *sigh* I like doing this at home better, anyway! I did not mean to mystify anything, and I apologize if it appeared that way. That list is specifically for my students; that's why there's some notation in it. Okay, I confess, I cut and pasted. Nevertheless, I believe that an informed patient who needs to make a rational choice about treatments should have some sense about statistics and what statistical processes mean. People always have agendas, and you need to be able to see past them. How will you do that without understanding the information they are using? You asked about specific points I listed. >>6. Do not confuse Prob(A, given B) with Prob(B, given A). This is probably the most mysterious if a person doesn't know probability notation. But the concept can also be challenging. Say you have some screening test for some dread disease, and when the results come back, the doctor says that she/he would like to do a second test. This is often because the first test came back positive. The question facing the patient at that point is: What is the chance that the patient has the disease, given that he/she has tested positive for the disease? One measure of the quality of a medical test is its sensitivity, which is defined as the chance that a patient tests positive for a disease, given that they have the disease. Compare these two questions: 1. What is the chance that a patient has a disease, given that he has tested positive for the disease? 2. What is the chance that a patient will test positive for the disease, given that he has the disease? Question 2 is answered by the sensitivity, while question 1 is the one I'd want answered when the doc tells me I tested positive. The probabilities calculated for each are almost never the same. But because they are similar sounding, people will often think they are -- even doctors. Similar fallacies in reasoning about medical tests involve false positive rates and/or specificity. My point to my students -- and to anyone who thinks of themselves as an informed patient -- is to keep these straight, and to not depend on a doctor to know the difference. >>7. Creating a confidence interval is like playing ring toss, not >>darts. Some on the NG may not know the term, but you've seen confidence intervals every time you've read a news account of a political poll that includes a margin of error. For example, a recent CNN poll gave President Bush a 31% approval rating by registered voters. The margin of error was +/-3%. What this means is that the actual percentage of voters who approve of the job he's doing is somewhere in the interval between 29% and 35%. Where it is precisely, we do not know and never will know. The 32% is our best guess, but we'd never expect that it actually is 32% -- 32% is not like a dart thrown at a bullseye. The actual percentage is like the peg in a game of ring toss, and the interval we create out of a sample is like the ring. If we've done a good job in creating the interval, we are pretty confident that it will contain the actual percentage -- the ring lands around the peg. (The ring toss metaphor is one I've used in class for a long time; I don't remember where I learned it. It's a pretty standard one in statistics courses.) I have only now started wading through the ADT literature -- I bought the new edition of Walsh today, and will start looking at the studies he cites as well as others brought up on this thread. But I'm guessing that every time a researcher claims that ADT (at whatever stage) prolongs life by so many months on average, that average comes with a margin of error. It's simply SOP in that type of research. An informed patient reading such studies should be able to interpret confidence intervals when encountered. And once again, I do not think you can count on doctors to do so. >>8. A hypothesis test always starts with a yes/no question, continues >>with one of many statistical processes, and ends with the probability >>that you are wrong if you answer "yes." But hypothesis testing is at the very heart of all of this! Every bit of research starts with a question, that a scientist then struggles to turn into something testable. That's the yes/no question. "Does starting ADT immediately upon a PSA rise prolong life?" After we've settled on the question, then the experiment, study, or trial begins to take shape, along with the statistical tests and the choices that go into it. The probability that we are wrong with you answer 'yes' is more commonly known as the p-value. The lower the p-value, the more convincing the result. You become a more efficient reader of scientific research when you understand this template. It just becomes easier to think about research critically. --charlie Signature6/2006 PSA 5.2 DRE suspicious 7/2006 Biopsy 2 of 10 positive Gleason 7(3+4) 11/2006 LRP Clear margins 1/2007 PSA < 0.01 3/2007 PSA < 0.01 6/2007 PSA < 0.01 so far, so good >>>6. Do not confuse Prob(A, given B) with Prob(B, given A). > [quoted text clipped - 16 lines] >1. What is the chance that a patient has a disease, given that he has >tested positive for the disease? For example, "my PSA is high - do I have the disease or just an inflamed prostate/BPH?). >2. What is the chance that a patient will test positive for the disease, >given that he has the disease? For example, "my biopsy came back negative - did the needle cores miss the cancer?" I can't see why anyone after a short browse through the archives here could possibly suffer the confusion implied by your (6). On the contrary, many newbies to this forum, after a positive biopsy and with a dash of pessimism nagging away, may very quickly graduate to a refinement of your (2) immediately above.."is my biopsy *really* only a gl(3+3) or did the needle miss some (3+4)?" We can quickly become quite sophisticated. >Question 2 is answered by the sensitivity, while question 1 is the one I'd >want answered when the doc tells me I tested positive. The probabilities [quoted text clipped - 6 lines] >informed patient -- is to keep these straight, and to not depend on a >doctor to know the difference. In the UK our first high PSA result would normally be reported and discussed by our GP. I might want to talk about the probability of a false positive (let's call it fp). Nowadays they're all online so he'll have the figures to hand even if not from memory. If it went to a biopsy, that result would be discussed by a urologist. If the biopsy were clear I might want to discuss the possibility of a false negative (fn). I don't see there's much scope for confusion. Nor do I see the point in equating fp=fn. You say these two values are "almost never the same". Why should they be? They arise out of radically different processes. One relates to the correlation between PSA values and prostate cancer, the other, to cite a parallel example, to the probability that a blindfolded man will fail to stick a pin in an apple precisely at the point where the fruit is bruised. I can be well aware of this distinction without knowing a dot about statistical method. >>>7. Creating a confidence interval is like playing ring toss, not >>>darts. [quoted text clipped - 24 lines] >studies should be able to interpret confidence intervals when encountered. >And once again, I do not think you can count on doctors to do so. Cripes! Do you really think that when we're told ADT will give us an extra x months/years we're banking on precisely that? Or even anything like that? Margin of error is most certainly a common sense concept and everyday life provides clear examples of varying magnitudes of margin. >>>8. A hypothesis test always starts with a yes/no question, continues >>>with one of many statistical processes, and ends with the probability [quoted text clipped - 8 lines] >probability that we are wrong with you answer 'yes' is more commonly known >as the p-value. The lower the p-value, the more convincing the result. Quite! That last sentence is fine. Why cloud the issue with your (8)? - the language, bordering on esoteric gobbledygook appears designed to mystify, not clarify. >You become a more efficient reader of scientific research when you >understand this template. It just becomes easier to think about research >critically. I disagree. I think you've dressed up some otherwise quite straightforward ideas in technical language. Of course, there will be some whose heads will spin when trying to read the results of a study but most here will have as clear an idea as is possible to glean of what is going on. Generally, I don't accept that your focus on the results and studies we see here is any the sharper for your expertise. I would though most certainly respect your insight on a flawed statistical study though I wouldn't put it past some on this forum to be as ready and able to blow the whistle. The devil is in the detail. Spotting the detail IMHO is not always in the gift of statisticians. > Nor do I see the point in equating fp=fn ....after a re-read - of course you didn't say fp=fn. Better would have been:- "nor do I see any reason why anyone would equate fp with fn" > Cripes! Do you really think that when we're told ADT will give us an > extra x months/years we're banking on precisely that? Or even > anything like that? Margin of error is most certainly a common sense > concept and everyday life provides clear examples of varying > magnitudes of margin. With ADT in particular, the survival benefit margin of error is huge and lopsided, ranging from zero to years (the negative survival benefit once reported was legitimately explained away). *However*, unless we have a medical reason to expect our own results will run long or short, it makes sense to assume -- i.e., bank on -- the median. I.P. >> Cripes! Do you really think that when we're told ADT will give us an >> extra x months/years we're banking on precisely that? Or even [quoted text clipped - 9 lines] > >I.P. Yes it does make sense, but there's a personal dimension to appraising data isn't there? It shouldn't be subjective but I suspect it is. Bank on, count on, expect, hope for, hope for better than...where a margin is huge there's opportunity a-plenty for pessimists and optimists to do their own thing. Someone heavy on the positive thinking can run his finger along the figures and pick one at the higher end - forget the median. Optimism transcends cold rationalism. >>> Cripes! Do you really think that when we're told ADT will give us an >>> extra x months/years we're banking on precisely that? Or even [quoted text clipped - 16 lines] > thinking can run his finger along the figures and pick one at the > higher end - forget the median. Optimism transcends cold rationalism. That little paragraph runs deep. Its opening and closing statements, while loaded with implications, both conflict and complement one another. On the complementary side, they give each of us room to dream, allowing dreamers to defend almost any interpretation they wish. On the conflicting side, the closing sentence implies optimism is better than rationalism, which is like saying apples are better than ice cream. One man's focus may legitimately emphasize *quality* of life over the next few years, while the next man may very clearly favor sacrificing quality of life for the slim chance that it might stack on a significant *quantity* of life at the tail end. Both are legitimate individual choices, and I submit that neither optimism nor rationalism transcends the other except as a basis for an individual choice. Even then one will seldom know whether a choice based on longevity was right, because he will seldom know whether it was treatment, lack thereof, or global warming that determined his funeral date. The one thing most of them *will* know once they're on it, if they pursue ADT, is how it affects their near-term QOL. I.P. Hi Charlie, I am a chemist and you forgot more about statistics that I would ever know or learned, so when the authors of a reference state the following, how do you interpret the results? 1. There is concern regarding the accuracy and completeness of information on HT and ADT from Medicare claims. 2. It is not possible to generalize the findings in this study to all men treated for locoregional CaP, since our sample was restricted to men aged greater than or equal to 65 who were not members of HMO and had both Medicare Part A and Part B coverage. Furthermore, there was no information on the dosage of HT and the combinational HT, androgen suppression in the form of orchiectomy and/or LHRH-Agonist plus anti- androgen such as flutamide or bicalutamide used in recent RCTs to examine the efficacy of ADT in CaP in terms of survival benefits. 3. There was no information on androgen-sensitive tumors (tumor markers), which, if available, will provide a better assessment (since CaP is initially androgen-sensitive, becoming insensitive with tumor progression to metastasis), for the effectiveness of ADT in CaP management. 4. The results in this study are unlikely to be due to bias such as measurement error, selection bias or confounding effect of other covariates on the effect of ADT on all-cause mortality. Selection bias is a major issue in observational studies including this present study. Older men in our sample who received ADT were different in spectrum of prognostic indicators than those who did not. Men who received ADT were less likely to be younger, married, educated, have low Gleason score, receive RP, RP/XRT, and more likely to have comorbid conditions and undergo watchful waiting (observational management). These factors are either associated with tumor virulence or poor outcome. This selection bias may very well have influenced the result of our study. However, we minimized some of these biases by the propensity scores, stratified by the primary therapy and adjusted the hazard ratio for the effect of ADT on mortality by these sources of noncomparability (confoundings). Nonetheless, like in all epidemiologic studies, our results might still be influenced by unmeasured and residual confoundings. 5. A statistically significant difference in these covariates was observed between those treated and untreated with ADT (2; P<0.05). However, after adjustment for propensity score to receive ADT, there was no statistically significant difference in comorbidity and SEER geographic areas in cases treated and untreated with ADT (2; P<0.41) and (2; P<0.39), respectively. Source: Prostate Cancer and Prostatic Diseases advance online publication 8 May 2007; doi: 10.1038/sj.pcan.4500973 Effectiveness of androgen deprivation therapy in prolonging survival of older men treated for locoregional prostate cancer L Holmes Jr1,2, W Chan3, Z Jiang2 and X L Du2 Thanks, RalphV www.pcainaz.org/phpbb <SNIP> > I have only now started wading through the ADT literature -- I bought the > new edition of Walsh today, and will start looking at the studies he cites [quoted text clipped - 18 lines] > 6/2007 PSA < 0.01 > so far, so good > Hi Charlie, > I am a chemist and you forgot more about statistics that I would ever > know or learned, so when the authors of a reference state the > following, how do you interpret the results? SNIP These are all valid questions, and each can be interpreted more than one way. But I think they're a dominant kind of tree in a forest of many kinds of trees, the next largest kind being the "Our study got solid, clear results; sorry if they disagree with all the other studies." What I'm trying to describe is a pattern of widely varying results obfuscated even further by a lot of mathematical CYA weasel words, and the forest's bottom line is that the medical community has no really useful quantifiable handle on ADT's benefits. The gurus argue over it endlessly, we argue over it endlessly, and in the mean time each cancer case does its own damn thing to the extent that our hair-splitting produces only a meaningless mean or median number plus a lot of unnecessary angst. Who is any of us when even guys like Partin and VA statisticians and meta-study researchers, with whole professional staffs cranking numbers for them, *still* have only widely varying *opinions* on the subject? I.P. Hello, Ralph: Well, thank you, but please understand that I am a mathematician who teaches at a small liberal arts college, and so I've had to learn statistics in order to teach the courses in stats and probability that we offer. I can hardly be called a statistician, with only one technical report under my belt and a handful of consultations with colleagues at the college who needed help wading through new things. I am approaching this in the same way. You do have me at a disadvantage, too, since you have the entire paper before you, and I have only the points you raise. My comments might sound a little trite therefore. But I will do my best, and certainly will include the Holmes paper on my "to do" list. > 1. There is concern regarding the accuracy and completeness of > information on HT and ADT from Medicare claims. I don't know exactly what the concern would be, as I know little about the way that Medicare is handled. I suppose that docs might want to pad claims? There might also be a long time lag between the time of, say, a Lupron shot and the time it's reported on a claim. Perhaps if the date the shot was administered is not required on the claim? But that doesn't seem reasonable. If a patient is shuttled between docs, that could also be a concern. > 2. It is not possible to generalize the findings in this study to all > men treated for locoregional CaP, since our sample was restricted to [quoted text clipped - 4 lines] > androgen such as flutamide or bicalutamide used in recent RCTs to > examine the efficacy of ADT in CaP in terms of survival benefits. The author appears to be reluctant to extrapolate the findings to a larger population. A standard caution. I'd also assume that the breakdown of the type of HT would be of extreme interest to researchers. Were I to read this in a conclusion of a study, I'd probably figure that a survival analysis based on the type of HT would be very interesting to those in the field. > 3. There was no information on androgen-sensitive tumors (tumor > markers), which, if available, will provide a better assessment (since > CaP is initially androgen-sensitive, becoming insensitive with tumor > progression to metastasis), for the effectiveness of ADT in CaP > management. There's a lot of cancer biology that I still need to learn, but I am beginning to wonder if this is the best way to put it? My understanding is that the tumor starts off with both androgen-dependent and androgen-independent cells. HT starves the dependent cells, clearing the path for the independent ones. What's in point 3 sounds like something different. But this could just be my ignorance. Again, I think this is a call for more research. > 4. The results in this study are unlikely to be due to bias such as > measurement error, selection bias or confounding effect of other [quoted text clipped - 13 lines] > epidemiologic studies, our results might still be influenced by > unmeasured and residual confoundings. Selection bias is due to the way the samples were selected; that's why the paper includes that discussion about the nature of the men in the sample. The author believes that they minimized the selection bias effects by stratification -- that essentially means that instead of one broad comparison, they probably did several smaller comparisons, grouping men with similar characteristics together -- compare younger men to younger men, older men to older men. I'd probably be able to say more about this if I knew the procedures they used. > 5. A statistically significant difference in these covariates was > observed between those treated and untreated with ADT (2; P<0.05). > However, after adjustment for propensity score to receive ADT, there > was no statistically significant difference in comorbidity and SEER > geographic areas in cases treated and untreated with ADT (2; P<0.41) > and (2; P<0.39), respectively. The first p-value says there was an effect. I am having a harder time interpreting the second -- again, I think it's a context thing. I think that the second is a test for the selection bias from point #4. In other words, the difference between the men treated or untreated is not explained by other factors, such as comorbidity or locale. I'll think about this some more. --charlie Signature6/2006 PSA 5.2 DRE suspicious 7/2006 Biopsy 2 of 10 positive Gleason 7(3+4) 11/2006 LRP Clear margins 1/2007 PSA < 0.01 3/2007 PSA < 0.01 6/2007 PSA < 0.01 so far, so good > >...Which elements of your stats course do you think would be of benefit? > [quoted text clipped - 13 lines] > through the course, we often return to these and see how we can > sharpen those questions. One area that news reporters often report in a misleading or wrong way is medical test results. In this 5-part series on PSA Screening: http://palpable-prostate.blogspot.com/2007/05/psa-screening-and-early-detection- part.html part 5 discusses actual examples of news and journal articles that relate to prostate cancer including: - one example that discusses the ECPA-2 prostate cancer test based on a news article from a major news organization - three examples that discuss the Miraculins prostate cancer test - one example that discusses the PSA Nanotest - one example that discusses color doppler It should be clear after reading this that if you do not have an appreciation of these concepts that one can easily be misled by a casual reading and that publication by a major news organization is no guarantee of good reporting. Parts 4 and 5 give the necessary background needed to understand testing concepts so read those at least (and preferably all parts) prior to reading part 5 where the actual examples listed above are located. --- The Palpable Prostate http://palpable-prostate.blogspot.com and discusses the results making use of medical testing concepts. It does require high school math to understand. >It should be clear after reading this that if you do not have an >appreciation >of these concepts that one can easily be misled by a casual reading >and that >publication by a major news organization is no guarantee of good >reporting. Why not simply short-circuit the risk and read the report rather than the reported report? > >It should be clear after reading this that if you do not have an > >appreciation [quoted text clipped - 5 lines] > Why not simply short-circuit the risk and read the report rather than > the reported report? Reading the primary literature is a good idea but is not an alternative to understanding and critical thinking. --- The Palpable Prostate http://palpable-prostate.blogspot.com >> >It should be clear after reading this that if you do not have an >> >appreciation [quoted text clipped - 12 lines] >The Palpable Prostate >http://palpable-prostate.blogspot.com Understanding and critical thinking in all the arts and disciplines is a laudable objective but this is a PCa forum. Can you cite a PCa study where the results are expressed in terms that require anything beyond basic numeracy to be *fully* understood? > Can you cite a PCa > study where the results are expressed in terms that require anything > beyond basic numeracy to be *fully* understood? I have taken many undergrad and very heavy gradate courses in probability and statistics, Bayes' risk, stochastic noise theory and applications, etc. At one time I could write and/or solve page-filling equations in the stuff; Norman Thagard and I used to sit up way into the night discussing the stuff in depth. I've forgotten a lot of it by now, but not all of it, and my textbooks and homework are in a filing cabinet across the room. Yet this statistical $#!+ in *most* of these studies still glazes my eyes over, on two levels. One, I'm *still* not used to the nomenclature they use. Two, their weasel-worded CYA disclaimers obfuscate most of most studies' bottom lines. Classic recent example: "The results in this study are unlikely to be due to bias such as measurement error, selection bias or confounding effect of other covariates on the effect of ADT on all-cause mortality. Selection bias is a major issue in observational studies including this present study. Older men in our sample who received ADT were different in spectrum of prognostic indicators than those who did not. Men who received ADT were less likely to be younger, married, educated, have low Gleason score, receive RP, RP/XRT, and more likely to have comorbid conditions and undergo watchful waiting (observational management). These factors are either associated with tumor virulence or poor outcome. This selection bias may very well have influenced the result of our study. However, we minimized some of these biases by the propensity scores, stratified by the primary therapy and adjusted the hazard ratio for the effect of ADT on mortality by these sources of noncomparability (confoundings). Nonetheless, like in all epidemiologic studies, our results might still be influenced by unmeasured and residual confoundings." In English: Objects in the mirror be imaginary. I.P. >> Can you cite a PCa >> study where the results are expressed in terms that require anything [quoted text clipped - 33 lines] > >I.P. Yes, I spotted that chunk earlier. Instead of *fully*, I should have said "fully". > >> >It should be clear after reading this that if you do not have an > >> >appreciation [quoted text clipped - 17 lines] > study where the results are expressed in terms that require anything > beyond basic numeracy to be *fully* understood?- Hide quoted text - I already cited a post on my blog that discusses several examples. Note that only some of them refer to news articles. Others refer to the medical literature. --- The Palpable Prostate http://palpable-prostate.blogspot.com >Can you cite a PCa >> study where the results are expressed in terms that require anything [quoted text clipped - 3 lines] >Note that only some of them refer to news articles. Others refer to >the medical literature. I can't see a single item that justifies an insider's knowledge although there's unquestionably some advanced material on the blog which for those so inclined would be interesting in itself. I'm talking here about what we --need-- to know to make as informed a decision *as possible*. As we see from the continued controversy over the ADT issue, a definitive treatment conclusion is elusive because it's dependent on weighing up SEs against survival benefit, one a value judgment the other, as far as can be gauged from the stats available to date, might as well be. Digging deeper into the methodology of an arguably flawed study doesn't get us any closer to the subjective root of the dilemma. > The evidence in a survival advantage is > there for all to see and it is not just a few months depending on the > severity of the disease at diagnosis and on the treatment applied. Do > not be fooled by misinterpretations on statistical results. Are we to assume, then, that Johns Hopkins and Walsh peg the post-RP ADT survival advantage at zero because they "got fooled"? I.P. On June 27, Mike Freely replied to Ralph V: > Are we to assume, then, that Johns Hopkins and Walsh peg the > post-RP ADT survival advantage at zero because they "got fooled"? I wonder whether anyone here can and will quote *exactly* what Walsh wrote at page 479 et seq. of his latest book? I would rather see the primary source, rather than what someone claims he says. What was Walsh's source material? Clinical trials, or his personal experience with his own patients? How many in the cohort? What was their clinical status in the beginning? And at the end? And what, exactly, was selected as the end point? And so on, leading to validation (or not) of whatever it is that he actually wrote? Regards, Steve J And regarding waiting for symptoms to manifest: "There is NOWHERE in oncology where waiting for the tumor cell population to increase (and to mutate) is in the better interests of the patient." (emphasis in original) --Stephen B. Strum, MD > On June 27, Mike Freely replied to Ralph V: > [quoted text clipped - 10 lines] > selected as the end point? And so on, leading to validation (or not) of > whatever it is that he actually wrote? All valid questions, addressed to various degrees in the book, which anyone making decisions should read, IMO. OTOH, the body of ADT knowledge, including the huge VA data he references in his ADT discussion, were sufficient to convince Walsh, which goes a long ways towards convincing me at least until my next decision makes me dig deeper. I absolutely concur that no one should take anyone else's comments about a book too seriously; it's vital that each person read such sources first-hand to make sure its conclusions and opinions apply to him. For example, Messing's survival data refer only to RT pts, Walsh's pronouncement applies only to RP pts, and I don't know *what* Lam's unheard-of 10-year claim refers to. I.P. Messing's results apply to RP patients with lymph node involvement treated early versus delayed treatment. Results after 11.9 years (range, 9.7-14.5 years) in . Immediate (n = 47) vs delayed (n = 51) androgen deprivation following radical prostatectomy are: Immediate therapy Total number of deaths: 17 (36.2%) Deaths from prostate cancer: 7 (14.9%) Delayed therapy Total number of deaths: 28 (54.9%) Deaths from prostate cancer: 25 (49.0%) These figures are clear and after almost 12 years some of these men dodged the bullet and avoided dying of PCa... RalphV www.pcainaz.org/phpbb > For example, Messing's survival data refer only to RT pts, Walsh's > pronouncement applies only to RP pts, and I don't know *what* Lam's > unheard-of 10-year claim refers to. > > I.P. > Messing's results apply to RP patients with lymph node involvement > treated early versus delayed treatment. [quoted text clipped - 13 lines] > These figures are clear and after almost 12 years some of these men > dodged the bullet and avoided dying of PCa... So can anyone explain how/why Walsh disputes that? I.P. Because for years and years he has been opposed to androgen deprivation AND because he has no randomized controlled clinical trial on his own to disprove Messing's results. He claims that the study lacks statistical power ignoring the results of Zincke studies that show the same benefit. Then, Dr. Walsh is a great surgeon and has a healthy ego... RalphV www.pcainaz.org/phpbb > So can anyone explain how/why Walsh disputes that? Years ago when Dr. Walsh was critical of Dr. Messing's study he editorialized his reasons. At that time, 6 or 7 years ago I wrote the following that might answer your question in a more proper way. Dr. Walsh, to my knowledge, has not answered the latest update results: Hello All, The unfavorable comments about the Messing study are of course coming from Dr. Walsh who is known to be openly opposed to hormone suppression. Let's look at the critical comments: PW: 1) Such treatment, traditionally reserved for men with metastatic disease, results in major objective and subjective benefits in most patients. However, in approximately 50 percent of patients, disease progression occurs 12 to 18 months after the initiation of treatment, and as a result, survival rates have not increased over the past five decades. RV: Neither 50% of patients fail at the initiation of therapy or it happens in 12 to 18 months. Survival rates have increased in the past decade as a result of detecting earlier stages of advanced disease and application of early hormone suppression treatment. Clearly the argument is outdated. Circa 1960 and very misleading. Survival as measured by a reduced PCa mortality rate is known and documented. PW: 2) In an ongoing study by the European Organization for Research and Treatment of Cancer, 302 patients with stage D1 disease (which includes nodal but not extranodal metastases) who did not undergo radical prostatectomy were randomly assigned to receive either immediate or delayed androgen-deprivation therapy. With a median follow-up of six years, no difference in cancer-specific or overall survival has yet been found (Schroder FH: personal communication). RV: Here Dr Walsh uses a comparison to a study that is about more advanced disease, unpublished results, using some personal communication with the investigator, and in which surgery was not used to cast a shade of doubt on the Messing study. Messing treated early stages of advanced disease. The EORTC study is another ball of wax which eventually will answer the question in that set of patients. Apples and oranges and one more smoke screen... PW: 3) The Mayo Clinic series, which represents the largest retrospective series, with a non randomized control group and almost three decades of follow-up, found that the survival advantage in favor of immediate androgen-deprivation therapy was limited to DNA diploid tumors and became apparent only after 10 years. RV: The Mayo Clinic has been advocating early intervention with hormonal suppression after advanced disease is found at surgery. Here Dr. Walsh admits that there is a survival advantage to debulking and early suppression but he adds, it only happens in diploid disease and only after ten years. Aneuploid disease is a progression stage from diploid disease. One more reason to initiate early suppression to avoid the natural progression of the disease from diploid to aneuploid. PW: 4) . One concern is that the study never realized its projected goal of 240 patients. This is important, because the outcome of patients with nodal metastases is extremely variable and can be affected by a number of known and possibly unknown prognostic factors. (7,8,9) Such effects on the outcome of trials can be minimized by randomization, but the study by Messing et al. was relatively small and might have been affected by imbalances of factors that had not been identified at the time the study began. One factor that might have influenced the outcome is the lack of a central pathological review to assess the Gleason scores. (5,7,8) The absence of a correlation between histologic grade and survival suggests that an imbalance could have been present but unrecognized. RV: Dr. Walsh uses the argument that the outcome of patients with nodal disease is extremely variable as a probable cause for the survival advantage in the study . When he compares Messing results to all other studies in this Editorial, the same variability exists, but there he chooses to ignore this variability in the other studies mentioned. Then he proceeds to chastise the lack of central pathological review (consistency) in a multi center, randomized controlled trial as a potential cause of the survival advantage. I know that in some of his own multi center studies there is no central pathological review either...but conveniently he ignores that. RV: Bottom line is that earlier stages of the disease more often respond to hormone suppression because the bulk of the cancer is more androgen dependent. The lower the tumor burden the more effective and prolonged the response. You have to ask yourself why when we are always insisting in proving things with randomized controlled trials and there is a world of evidence that hormone suppression is more effective when tumor burden is low, we come up with all these critical reason to explain results we don't agree with even when it is a randomized controlled trial? Those that are critical of this study do not have a clinical randomized controlled trial to prove that their own method of treatment improves survival by one day...but they believe it does and continue to do it. I think they should talk about proving themselves first! RalphV www.pcainaz.org/phpbb > So can anyone explain how/why Walsh disputes that? > Years ago when Dr. Walsh was critical of Dr. Messing's study he > editorialized his reasons. At that time, 6 or 7 years ago I wrote the > following that might answer your question in a more proper way. <SNIP> Thank you very much for a substantive, topical response. It is infinitely more useful than your previous OT personal blast on Walsh's bias and ego in place of this far more topical discussion. Ever meet a doctor, especially a surgeon, without an ego? Earlier, you advised Rosbif to avoid being misled by misinterpretation of statistical results by: > 1. Learn statistics. > 2. find someone you know and trust, that understand them and explains > the results to you. Exactly what Walsh's book does, far better than most other books I've seen. When my PSA next returns and doubles, presenting another decision, I'll be studying that chapter closely. > 3. Until you do 1 or 2, doubt the interpretations given here... > more so when they claim to have done extensive research. And on what is any medical knowledge based if not research, the more extensive the better? > The unfavorable comments about the Messing study are of course coming > from Dr. Walsh who is known to be openly opposed to hormone > suppression. Actually Walsh very strongly recommends, even insists upon, its use in several scenarios, especially with initial RT or upon bone mets and also in more specific cases. Considering that so many PC books include only paragraph or a page on ADT, I find Walsh's 40-page discussion (expanded maybe 25% from his previous book) of ADT from many perspectives very insightful. IMO, it's a must-read place to start one's research on the topic, particularly post-RP (he says it's a no-brainer with RT). He describes 3 or 4 different ADT position camps and makes it very clear which one he resides in and why. Then he addresses much of what you mention here, plus much more, including why he puts so much faith in the old but still-valid (he explains) VA CURG study both his "Guide to Surviving Prostate Cancer" books cover. >> So can anyone explain how/why Walsh disputes that? I am at a disadvantage not having read Dr. Walsh new edition. Hopefully there he has modified his position on this issue. Patient benefits should be above everything else including old established perceptions. One fact that always amazes me is that invariably those opposed to early androgen suppression are fast to point all the side effects of androgen deprivation, but fail to recognize that allowing prostate cancer to progress is not without consequences. More so when dealing with early and aggressive recurrence after localized therapy. The assumption is that as PSA increases, symptoms fail to appear or are of no consequence, if they do. Save it for a rainy day! Wait till it shows in bone so that they can then treat it. Sure, at that point, Dr. Walsh and others figure that hormone suppression is necessary and effective. What is ignored is that invariably disease control and response is not as good as when there is less disseminated disease and tumor volume. This is well known since 1989 when Dr. Crawford did his seminal study. In the past, I perceive a lack of interest about this topic in Dr. Walsh's comments as he failed to recognize that urology is dealing with a new set of patients with different degrees of advanced prostate cancer in whom the use of hormone suppression (including intermittent) after failed localized treatment offers improved survival, disease control, avoidance of major complications and in general better quality of life than when the disease is allowed to progress unchecked and then it fails to be controlled with suppression at a later point in time. It is interesting to read some the hidden thoughts in a JHU study such as this one: "The overall 10- and 15-year metastasis-free survival rates in the present report were 87% and 82%, respectively. Zincke et al2 previously reported 10- and 15-year metastasis-free rates of 82% and 76% in more than 3000 men undergoing radical prostatectomy. In a multi- institutional study, Gerber et al6 reported 10-year metastasis-free rates that varied directly with tumor grade (low, 87%; intermediate, 68%; and high, 52%). When patients in our series were divided into these same categories, the metastasis-free rates at 10 years were better than those reported for those who had the low-grade tumors (100%) and intermediate-grade tumors (91%) but were somewhat lower in the higher-grade tumors (43%). This lower rate for the high-grade tumors may be due to the lack of early hormonal therapy in our patients with high-grade disease. Although this issue was not specifically addressed by Gerber et al,6 more than one quarter of the men in the report from the Mayo Clinic received either early hormonal or radiation therapy.2" Here they admit that the difference between studies and the better results in other institutions "may be" due to early suppression. Interesting, but they, for whatever reason, still do not support the notion... RalphV www.pcainaz.org/phpbb Source: Natural History of Progression After PSA Elevation Following Radical Prostatectomy Charles R. Pound, MD; Alan W. Partin, MD, PhD; Mario A. Eisenberger, MD; Daniel W. Chan, PhD; Jay D. Pearson, PhD; Patrick C. Walsh, MD JAMA. 1999;281:1591-1597. > Actually Walsh very strongly recommends, even insists upon, its use in > several scenarios, especially with initial RT or upon bone mets and also [quoted text clipped - 12 lines] > > - Show quoted text - >I am at a disadvantage not having read Dr. Walsh new edition. > Hopefully there he has modified his position on this issue. Patient > benefits should be above everything else including old established > perceptions. I used his First Edition almost as a bible, but lately have found it lacking due to new findings in the last half decade. I have commented that I would like to see his book in a revised form, but from the few descriptions I can find of it, I am not thrilled. If anyone has anything to say good (or bad) about the Second Edition, please let me know. > One fact that always amazes me is that invariably those opposed to > early androgen suppression are fast to point all the side effects of > androgen deprivation, but fail to recognize that allowing prostate > cancer to progress is not without consequences. And, as I like to add, the side effects are so varied as to be impossible to predict prior to actually taking the treatment. The people who tolerate the medication the worst seem to know it within months. And, none that I know of who have stopped within months had any lingering effects. SignaturePSA 16 10/17/2000 @ 46 Biopsy 11/01/2000 G7 (3+4), T2c RRP 12/15/2000 G7 (3+4), T3cN0M0 Neg margins PSA <.1 <.1 <.1 .27 .37 .75 PSAD 0.19 years EBRT 05-07/2002 @ 47 PSA .34 .22 .15 .21 .32 PSAD .056 years Lupron 07/03 (1 mo) 8/03 and every 4 months there after PSA .07 .05 .06 .09 .08 .132 .145 PSAD 1.4 years Casodex added daily 07/06 PSA <0.04, <0.05, <0.04 (06/12/2007) Non Illegitimi Carborundum > I am at a disadvantage not having read Dr. Walsh new edition. > Hopefully there he has modified his position on this issue. Patient > benefits should be above everything else If you would read it, you'd see why he generally opposes early asymptomatic post-RP ADT: he feels the clinical evidence proves it diminishes QOL without prolonging life, and backs up his conclusion with data. > One fact that always amazes me is that invariably those opposed to > early androgen suppression are fast to point all the side effects of > androgen deprivation Why else (other than its expense) would anyone oppose it? Its therapeutic index -- its ratio of benefits to SEs -- is the very basis of their objection, and the basis of my personal rejection of it. Understand I'm not against anyone else taking early ADT *I*F* they research the hell out of it first and make an informed decision; I'd take a hundred pills, bottles, and shots a day if they might help me and had zero downside. The only people I've seen recommend against anyone else taking early post-op ADT are Walsh and arguably PCRI's Scholz, who said since the outcomes are a wash, choose your adjuvant treatment by the SEs you prefer. It's precisely those SEs (cost notwithstanding) which drive Walsh, Scholz, et.al. and so many ADT recipients to reject early ADT or at the very least advise us to choose our treatment by SEs, not survival. > but fail to recognize that allowing prostate > cancer to progress is not without consequences. > invariably disease control and response is not as good as when there > is less disseminated disease and tumor volume. This is well known > since 1989 when Dr. Crawford did his seminal study. Among those disagreeing with that, citing SEs, QOL, and failure to prolong life, are: Dr. Patrick Walsh, PC treatment pioneer and former head of J-H's Brady Institute. The Mayo Clinic. Sloan Kettering. The ACS. Bubley (of Harvard) Several university oncology schools. A meta-study of all the ADT survival benefit literature (it found no evidence early adjuvant ADT provides any advantage, even with rising PSA.) http://npg.nature.com/pcan/journal/vaop/ncurrent/full/4500973a.html Quote: "ADT was not associated with significantly increased survival benefit for [64,475] older men with locoregional CaP" and the list goes on and on and on. Don't you think these people and institutions have seen all the references you and I could find, plus many more? There's no point in debating the issue here, or in having a reference citation contest, given the huge number of conflicting studies, the meta-study of which found no survival benefit. i.e., There *IS* no proof that satisfies the erudite list above of any early ADT survival advantage. That's good enough for me, and should persuade anyone considering early ADT to spend weeks researching it before bending over. Quoting Walsh: "No form of early hormone therapy delays progression of the disease. The only thing it delays is your *knowledge* of this progression." (emphasis his) I.P. The problem I am having with this exchange is that while I am talking at applying ADT early on when there is an established recurrence after treatment failure. You are talking about early ADT in cases when this is not still defined. This is probably happening, but not as a common occurrence as you imply. In 15 years in the support group movement I have never seen such cases. Exceptions would be those who are planning some form of radiotherapy and are in need of a gland volume reduction before treatment. Similarly for those who will not get near a scalpel or a linear accelerator, intermittent suppression a la Dr. Leibowitz is an option. Those I have seen and met. Dr. Bob's patients I have met for the most part are reasonably content with their decision although some had to do another round of suppression I might add.. I agree that the indiscriminate use of ADT in early disease is not supported by the medical literature. In all other instances where it is, involving recurrence, an early timing application has a disease- specific survival benefit along with a delay of disease progression. Decisions made based on that knowledge and being fully aware of the negative aspects of hormonal suppression, depend on the personal preferences of the patient. There's no point in debating the issue here, or in having a reference citation contest, given the huge number of conflicting studies, the meta-study of which found no survival benefit. i.e., There *IS* no proof that satisfies the erudite list above of any early ADT survival advantage. That's good enough for me, and should persuade anyone considering early ADT to spend weeks researching it before bending over. http://npg.nature.com/pcan/journal/vaop/ncurrent/full/4500973a.html RV: This study is a prospective execise of data extraction as compared to a randomized trial in which a treatment is properly examined. The autors write: " Older men in our sample who received ADT were different in spectrum of prognostic indicators than those who did not. Men who received ADT were less likely to be younger, married, educated, have low Gleason score, receive RP, RP/XRT, and more likely to have comorbid conditions and undergo watchful waiting (observational management). These factors are either associated with tumor virulence or poor outcome. This selection bias may very well have influenced the result of our study. However, we minimized some of these biases by the propensity scores, stratified by the primary therapy and adjusted the hazard ratio for the effect of ADT on mortality by these sources of noncomparability (confoundings). Nonetheless, like in all epidemiologic studies, our results might still be influenced by unmeasured and residual confoundings. This study also found that increase in overall mortality in this cohort was associated with advancing age, higher comorbidity score, higher Gleason score/AJCC stage, lower education and income, unmarried status and African American ethnic/racial group. These prognostic factors were significantly different between those who received HT and those who did not. These findings may be biased by differential distribution of these prognostic factors. While to some extent, this might be possible, it is however, unlikely that our finding of marginally significant increased risk of dying in those who received HT compared with those who did not was driven solely by this differential distribution of the prognostic factors, since we did minimize some of these imbalances by propensity scores, and adjusted for these confoundings in a multivariable Cox proportional hazard model using clinically relevant and statistically significant covariates. Likewise, there is a possibility of misclassification bias. SEER provided reliable information that was validated by SEER staff, and Medicare data have been both externally and internally validated by many researchers.33, 43, 44 However, we cannot exclude the possibility of misclassification of the factors that may have resulted from modeling categorized variables that were originally continuous and measured with nondifferential error.45" RV: Also notice that this study reports "overall" mortality and not disease-specific mortality. In the results the authors report: "However, after adjustment for propensity score to receive ADT, there was no statistically significant difference in comorbidity and SEER geographic areas in cases treated and untreated with ADT (2; P<0.41) and (2; P<0.39), respectively. RalphV www.pcainaz.org/phpbb > The problem I am having with this exchange is that while I am talking > at applying ADT early on when there is an established recurrence after > treatment failure. You are talking about early ADT in cases when this > is not still defined. Walsh and the rest of my list are talking about post-RP ADT any time before bone scans prove mets, or even the onset of symptoms. Clearly and long-rising PSA of 8.0 . . .fuhgheddaboutit. Add a hot spot on a bone scan . . . start paying closer attention, maybe thinkaboutit if you're otherwise healthy. Physical symptoms supported by lab results . . . compare the therapeutic index vs your life's priorities and decide which is the lesser of two evils . . . ADT or symptoms . . . 'cause your burial date will not change. > This is probably happening, but not as a common > occurrence as you imply. In 15 years in the support group movement I > have never seen such cases. The University of Washington Dept. of Oncology Tumor Board strongly advised me to begin ADT as soon as my negative-margins RP healed "Just In Case" my G-8 cancer was not eradicated and "because post-RP ADT is the normal, prudent course of action in all but the mildest cases." > In all other instances where it > is, involving recurrence, an early timing application has a disease- > specific survival benefit "The list" . . . Dr. Patrick Walsh, PC treatment pioneer and former head of J-H's Brady Institute. The Mayo Clinic. Sloan Kettering. The ACS. Bubley (of Harvard) Several university oncology schools. A meta-study of all the ADT survival benefit literature (it found no evidence early adjuvant ADT provides any advantage, even with rising PSA.) http://npg.nature.com/pcan/journal/vaop/ncurrent/full/4500973a.html Quote: "ADT was not associated with significantly increased survival benefit for [64,475] older men with locoregional CaP" and the list goes on and on and on. disagree. Your analysis of the "Nature" study I quoted above was enlightening (after reading it a few times), but its bottom line seems to me to be: "Yes, the ADT guys were older, sicker, and more medically ignorant than the non-ADT guys, which could explain their poor ADT response. But we did our very *damnedest* to methodically eliminate that bias and thus stand by our statement of limited, if any, survival advantage." > along with a delay of disease progression. On that they agree. > Decisions made based on that knowledge and being fully aware of the > negative aspects of hormonal suppression, depend on the personal > preferences of the patient. And on that, in theory, they concur. But we see individuals here and in the literature and whole studies showing how often pts are not offered that knowledge, awareness, and/or choice. If my objective here were to persuade you or anyone else for or against ADT at any phase, and if I had the time, I'd be citing references and counter-references with you until the cows come home. But that's not my objective. My objective is to make patients considering RP-adjuvant ADT aware that its survival benefit is far less clear-cut than, say, global warming, and that they would be well-advised to do their own research. They can swallow the fish (individual studies or opinions) people hand them and live or die by those data points, or they can learn to fish for themselves and make their own choices based on the far larger data base I presume The List bases its opinions on. I'm not qualified to debate the issue with you, but I must similarly assume The List is better qualified, in turn, than all of this forum combined. If the "No Proven Survival Benefit" camp were just a maverick or two, I'd shrug it off, but IMO it's too large and substantial to ignore. I.P. Comments under RV>+++++> > > The problem I am having with this exchange is that while I am talking > > at applying ADT early on when there is an established recurrence after [quoted text clipped - 9 lines] > is the lesser of two evils . . . ADT or symptoms . . . 'cause your > burial date will not change. RV>++++++++> Again, the thread started with Dr. Myers case. Positive lymph nodes and the aggressive course he took. You switch it to what you feel is your crusade against SEs induced by ADT. There is a major difference to apply ADT after recurrence and before recurrence. One is supported by the medical literature and the other is not at this point in time. > > This is probably happening, but not as a common > > occurrence as you imply. In 15 years in the support group movement I [quoted text clipped - 4 lines] > In Case" my G-8 cancer was not eradicated and "because post-RP ADT is > the normal, prudent course of action in all but the mildest cases." RV>+++++++> Again, you switch the topic to YOUR case. The reason they advised you is because GS 8 cases tend to recur more often after RP than lower Gleason scores. It was then your choice to do ADTor not. > > In all other instances where it > > is, involving recurrence, an early timing application has a disease- [quoted text clipped - 9 lines] > Bubley (of Harvard) > Several university oncology schools. RV>+++++++> Your list is meaningless as we are addressing different disease stages where ADT should or should not be applied. The following URLs provide a notion when ADT is applicable in the treatment of PCa. There are no recommendations or studies that support the notion that ADT should be used for biochemical failure (PSA alone). It is obvious that a treatment failure should raise a flag early on and patients should be aware of the possibilities. http://tinyurl.com/yt9cf8 (AUA Guidelines '07) http://www.cancer.gov/(Prostate Cancer PDQ) Nowhere in these sites, there is a recommendation to use ADT in early stage prostate cancer. They recommend ADT for high risk, poorly differentiated disease. > A meta-study of all the ADT survival benefit literature (it found no > evidence early adjuvant ADT provides any advantage, even with rising PSA.)http://npg.nature.com/pcan/journal/vaop/ncurrent/full/4500973a.html [quoted text clipped - 10 lines] > did our very *damnedest* to methodically eliminate that bias and thus > stand by our statement of limited, if any, survival advantage." RV>+++++++> The authors could have done their "damndest" but their results are statistically insignificant based on their reported p vaues. They extracted data from the SEER database that segregates GS as follows: A Gleason sum of 2, 3, or 4 is translated to a well-differentiated tumor; a sum of 5, 6, or 7 to a moderately differentiated tumor; and a sum of 8, 9, or 10 to a poorly differentiated tumor. The acceptance of a Gleason sum of 5, 6, or 7 to a moderately differentiated or intermediate-risk category can be a problem by mixing low-risk disease (Gleason sum of 5 or 6) with more aggressive disease (represented by a Gleason sum of 7). A Gleason sum of 7 can trend to intermediate or high-risk disease, depending on how much of pattern 4 is present in the biopsy. This is compounded by the absence of a discriminatory PSA result for patients in this study. > > along with a delay of disease progression. > [quoted text clipped - 18 lines] > themselves and make their own choices based on the far larger data base > I presume The List bases its opinions on. RV>+++++++> My objective is to stick to information pertaining to the thread and not what you want to address. This confuses the issue. If you want to discuss the validity of ADT for high grade disease, start other thread. > I'm not qualified to debate the issue with you, but I must similarly > assume The List is better qualified, in turn, than all of this forum > combined. If the "No Proven Survival Benefit" camp were just a maverick > or two, I'd shrug it off, but IMO it's too large and substantial to ignore. RV>++++++> I never intend to debate issues with you because you have the tendency to alter the thread to fit your purpose or to support your treatment decision. This has nothing to do with that. It has to do with the potential benefit of treating aggressive disease with a combination of aggressive treatments with the ultimate objective of defeating prostate cancer. Dr. Myers choice was his and might not fit others, but the intent was to demonstrate what a highly qualified person who happens to be a medical oncologist opted for treatment. RalphV www.pcainaz.org/phpbb > the thread started with Dr. Myers case. Positive lymph nodes > and the aggressive course he took. You switch it to what you feel is > your crusade against SEs induced by ADT. > My objective is to stick to information pertaining to the thread and > not what you want to address. This confuses the issue. If you want to > discuss the validity of ADT for high grade disease, start other > thread. On the contrary, here's a quote from your opening post in this thread: "[Meyers] like many of us undergoes androgen deprivation as early as possible in an effort to extend his survival. Is it possible that he misunderstands the voluminous medical literature indicating the futility and inefficacy of early or even hormonal suppression at all? I doubt that he would undergo such radical therapies when the evidence is so tilted to support the opposite. That is why I believe that the proof of the pudding is in the eating...The evidence in a survival advantage is there for all to see and it is not just a few months depending on the severity of the disease at diagnosis and on the treatment applied." > There is a major difference > to apply ADT after recurrence and before recurrence. One is supported > by the medical literature and the other is not at this point in time. So why does "The List" disagree with that statement? >>> This is probably happening, but not as a common >>> occurrence as you imply. In 15 years in the support group movement I [quoted text clipped - 5 lines] > RV>+++++++> > Again, you switch the topic to YOUR case. Nope. I used my case merely to present their broad statement in context. > RV>++++++> > I never intend to debate issues with you because you have the tendency [quoted text clipped - 5 lines] > but the intent was to demonstrate what a highly qualified person who > happens to be a medical oncologist opted for treatment. Then how did you segue from his personal choice to your blanket statement I quoted in my opening paragraph above? Ralph, you're playing the same game you accuse me of, and continuing to insist that you know more about ADT's benefits than Walsh, Mayo, Sloan Kettering, the American Cancer Society, Bubley, several university oncology schools, the meta-study of all the ADT survival benefit literature, and http://npg.nature.com/pcan/journal/vaop/ncurrent/full/4500973a.html . Folks, do your own research. Bye. I.P. > Exceptions would be those who are planning some form of radiotherapy > and are in need of a gland volume reduction before treatment. [quoted text clipped - 3 lines] > part are reasonably content with their decision although some had to > do another round of suppression I might add.. I understand that Dr. Leibowitz's conclusions have not been backed by rigorous trials/studies like some of the other physcians mentioned in this thread. At one time I seem to recall he was suggesting cures were attainable for "all" those completing his program.... > The problem I am having with this exchange is that while I am talking > at applying ADT early on when there is an established recurrence after > treatment failure. You are talking about early ADT in cases when this > is not still defined. This is probably happening, but not as a common > occurrence as you imply. In 15 years in the support group movement > I have never seen such cases. I had a urologist insist that I begin ADT ** BEFORE ** radiation. Worse mistake I ever made in my entire life, and I had only 1 injection before stopping because of active SE's and others reported by the FDA. >> The problem I am having with this exchange is that while I am talking >> at applying ADT early on when there is an established recurrence after [quoted text clipped - 7 lines] > Worse mistake I ever made in my entire life, and I had only 1 injection > before stopping because of active SE's and others reported by the FDA. Prior to RT is one of the ADT no-brainers even according to Walsh, as it does enhance RT's function. It's also of short duration, helping us gut out any SEs we encounter and rendering their likelihood of permanence to near-zero. I wonder how hard your uro tried to prevent or ameliorate your SEs or help alleviate your fears of the ones you didn't have. I.P. >> I had a urologist insist that I begin ADT ** BEFORE ** radiation. >> >> Worse mistake I ever made in my entire life, and I had only 1 injection >> before stopping because of active SE's and others reported by the FDA. > Prior to RT is one of the ADT no-brainers even according to Walsh, as it > does enhance RT's function. It's also of short duration, helping us gut > out any SEs we encounter and rendering their likelihood of permanence > to near-zero. I wonder how hard your uro tried to prevent or ameliorate > your SEs or help alleviate your fears of the ones you didn't have. He never once mentioned SE's before or after that one injection. It's kinda difficult to alleviate a fear of shortness of breathe or not being able to cross a street without a leg crumbling beneath both which never occurred before the injection. Or reading, after the fact, that the drug can cause a disease you already have. "Will it exacerbate my AS?" was the thought running through my mind. I stated rather firmly that I was NOT going to continue and he said, "OK." >>> I had a urologist insist that I begin ADT ** BEFORE ** radiation. >>> [quoted text clipped - 8 lines] > > He never once mentioned SE's before or after that one injection. Exactly why I and several book and study authors harp on SE awareness, and exactly why I can't understand why anyone takes it so personally when I discuss those in-house or more publicly documented SE awareness issues, especially given the relatively high number of very significant SE problems our small forum has personally encountered. > It's kinda difficult to alleviate a fear of shortness of breathe or > not being able to cross a street without a leg crumbling beneath > both which never occurred before the injection. Your SEs seem unique to your AS + ADT, but a) your uro should have prepared you for them and b) several of the normal ADT SEs are also frightening if not expected and/or dangerous if not treated aggressively. > Or reading, after the fact, that the drug can cause a disease you > already have. "Will it exacerbate my AS?" was the thought > running through my mind. If nothing else, your case plus Googling <ankylosing spondylitis adt> prompts me to add another recommendation for anyone considering ADT in the presence of any other "condition": Google the two together. I'm going to Google ADT with several words such as Caucasian, stubborn, heartburn, lipids, rednecks, and engineers when my PSA resurfaces. I.P. > I had a urologist insist that I begin ADT ** BEFORE ** radiation. > > Worse mistake I ever made in my entire life, and I had only 1 injection > before stopping because of active SE's and others reported by the FDA. That's usually how people find allergies to penicillin, latex, another elements of which some are more sensitive than others or who have grown an intolerance for. I would not consider your acquiescing to your doc a mistake. SRK |
Reply to this Message |
 Sign In
 Join
 My Latest Posts My Monitored Threads My Blog My Photo Gallery My Profile My Homepage Start New Thread Enable EMail Alerts Rate this Thread
|
©2008 Advenet LLC Privacy Policy - Terms of Use
This website includes both content owned or controlled by Advenet as well as content owned or controlled by third parties.