pathology

Frequently on this forum individuals are encouraged to obtain a second opinion on pathology slides. This practice in general is a reasonable suggestion and I have made this very recommendation on a number of occasions. Very often when this recommendation is suggested, specific names of pathologist are suggested.

It is my understanding that tissue identified as difficult to provide an accurate pathology interpretation on (such as prostate tissue, needle breast biopsy, and some others) have a mandatory prospective pathology review by another pathologist to confirm the interpretation. This peer review is performed even if the original interpretation was negative or positive for a cancerous condition. So in an accredited pathology lab all of these difficult tissues have had at least 2 individuals come to an agreement on the interpretation before a final report is issued. In addition to the aforementioned prospective review, all accredited pathology services also have a required retrospective review of a significant number of final reports as part of an ongoing quality program.

So the suggestion that all the wonderful academic facilities that we have in the United States designated as centers of excellence consistently fail to provide an accurate assessment of prostate pathology seems to be more of an urban legend than a scientific fact.

So I ask this of the contributors to this forum that I have great respect for because they have a personal standard of providing only information that is peer reviewed and is accepted as a standard of care…where is the evidence to support this recommendation… and where is the evidence that identify these specific providers as the undisputed foundation of prostate pathology truth?

Hero

Who the heck knows, Herophilus..your points are well taken, though. However, there are many recommendations made on this forum based on limited experience, their own, or looking at many studies, but none with the day to day contact an MD has, up close and personal with prostate cancer cases. Which is not to say personal anecdotes are not useful and certainly review of "peer reviewed papers" has some credence. But exactly how much to the individual? I saw not to long ago a comment on a forum by a participator that said
"I am hesitant to make a diagnosis without more information".. Really??? As a cancer patient and forum member , he feels qualified to make a diagnosis?

Good point, Sheepguy.

I was concerned that a resident read my husband's biopsy and maybe it should be sent to Johns Hopkins. Our internist assured us that the attending pathologist reads all the slides after the residents do, and that the pathologist is first rate.

It's MassGeneral and at some point you have to have a little faith. We both love Johns Hopkins, but were OK keeping his slides in Boston.

JT, my dad as very qualified as he was, pronounced the 351 Cleveland as in a class by itself. I'm sure that lots of 350 Chevy guys felt the same way in 1970. Your statement failed my ask... No Scientific basis.

Hero,
Would you have your airplane engine rebuilt by your local Chevy dealer? It's common sense that a specialist is better than a generalist in most instances.

Ha ha @ "IS the bank" on Epstein.

Stll, it appears to have been higher grade. If I'm not mistaken Pratoman your final pathology was upgraded from eptstein's.

sheepguy said...
The same skill "curve" then applies to the specialist prostate cancer readers as well. It too, is a skill unto itself. It reminds me of the assumption that if a surgeon has don 1000 proceedures, he is good..not necessarily, he may have been doing it poorly 1000 times not all people migrate to the far end of the "skill curve" with experience.

@Sheep, of course there is no assurance that any one surgeon is better because of experience. However, studies have shown that experience does matter in choice of a surgeon and, if one wants to improve his odds of success, choosing an experienced surgeon is one way to do it. Here is something I posted here on this subject in 2010:

The importance of surgeon's experience as it relates to prostate cancer outcome is underscored by the results of a study reported in the Journal of the National Cancer Institute (volume 99, page 1171).

Researchers analyzed the outcomes of 7,765 radical prostatectomies performed by 72 surgeons at four major academic medical centers. "Biochemical" recurrence was defined as a postsurgery PSA level greater than 0.4 ng/mL followed by a subsequent higher PSA level. The analysis took into account patient and tumor characteristics, such as pre-operative PSA level and Gleason grade. The men's PSA levels were measured every three to four months in the first year after surgery, twice in the second year, and annually during the following years.

The researchers found that surgical outcomes improved along with the number of radical prostatectomies a surgeon had performed, leveling off only after about 250 surgeries. The five-year probability of experiencing a recurrence of prostate cancer was 18% for surgeons who had performed only 10 operations compared with 11% for surgeons who had performed at least 250 surgeries.

Bottom line on prostate cancer surgery: The results suggest that you can improve your odds of a successful outcome from radical prostatectomy by taking time to find a surgeon with extensive experience.

@Hero, I think the stats quoted by Allen are pretty revealing. Also, if you want to Google the subject, you can see that determination of PCa pathology on a biopsy is a very difficult task despite the guidelines provided. Allen's stats, Prato's experience (and many more like it) and common sense are enough for me to continue to suggest that new members seek an opinion from a pathologist who specializes in reading PCa pathology.

Jim

This topic is of great interest to me as I was upgraded by JH from Gleason 8 to Gleason 4+5. The original classification was done by a pathology lab that I don't know much about except that it is a large lab used a lot by my uro and is located in the DC area.

In case of Gleason 8 - 10 it looks like the study that TA mentioned found by Epstein that 30% were misclassified as lower grades. I would seem to be in this category. So here is the problem: the only reason the grade matters is that in most studies it is the most important classifier that determines prognosis and the various success rates. If most patients are not getting their Gleason score read at JH and the score is somewhat subjective then in most studies the Gleason groups are shifted lower (as being downgraded from 9 or 10 to 8 by JH does not appear to be very common). So if I am a person that was given 4+5 by JH should I be looking at the results for 4+5 or actually for 4+4, as that group likely contains many people with similar cancers to me that did not send their slides to JH? This may seem academic to some but not to those like me because rates for 4+4 when teased out are always much better than for 4+5.

So here is a PhD thesis idea: an objective Gleason scoring method, one that does not depend on having one of three people in the world read your slides. Or if that is too difficult, have one of those 3 people read every set of slides in every major study that is to be taken seriously.

Sorry for the rant but this drives me crazy.
Tom

Sheepguy, exactly my point.

I forgot to mention that they updated the Gleason scoring system in 2005 so long range studies that follow men for 15 to 20 years would also likely be affected by this change. Most significantly for high Gleason scores they changed the way the tertiary grade is treated. In my case, as far as I can determine, this is the reason for the upgrade.

I wish there were a computer program that could accurately read a biopsy core, but unfortunately, it takes judgment honed by lots of experience. As hard as it is to read a Gleason grade, reading an mpMRI accurately involves even more judgment and experience. It worries me that everyone is rushing to put them in their facilities, but there are few radiologists who can read them well. There's a whole lot of misreading going on. I cringe whenever I hear in my groups that someone had a CDUS reading by someone (e.g., Scholz) that is a relative novice at it.

BTom - I understand your point. The risk stratification is based on the Gleason score and staging they thought was there (including misreads, biopsy sampling misses, and imaging limitations) and not what was really there on final post-surgery pathology. All we can comfortably say is that the clinical diagnosis is highly correlated with the pathological diagnosis. It's greatest importance is in the treatment decision and the risk of under or over-treatment. That's why I always contend that cT3 is a more serious finding than pT3 - if it's big enough to be clinically apparent, it's big. Similarly, considering the prevalence of undergrading, GS8-10 found on biopsy is especially serious.

The 2005 changes had an interesting effect. It's called the "Will Rogers Effect" based on his joke "When the Okies came to California, the average IQ of both states went up." The 2005 changes mostly shifted some of the riskier GS 6s into the GS 7 category, causing the average risk of both categories to go down. A similar effect is now occurring with the use of 3T MRIs for staging (aside from misreadings); it's shifting T2s into the T3 category thereby causing the risk associated with both categories to go down. Risk stratification systems have not yet incorporated this.

- Allen

It would be nice if this conversation wasn't required, wouldn't it? It's a reminder to me that medicine is as much of an art as it always has been. Judgement is required. The pathologist exercises judgement, as does the radiologist (exactly how many seeds? where do I put them?) and the surgeon (how wide a margin to I take?). We can improve the odds by going for high levels of experience and specialization, but in the end, it's not a sure thing...and likely never will be.

You can have 10 independent readers, but if none of them are really expert, what good is it?