Toward a new prostate cancer grading system — step 1


In the most recent issue of AUA News, Dr. Jonathon Epstein of  Johns Hopkins — unarguably one of the world’s leading prostate cancer pathologists — has laid out the general principles behind a new, validated, prostate cancer grading system that is designed and intended to replace the traditional Gleason grading system over the next few years.

The article on page 23 in the June issue of AUA News is a written summary of a presentation made by Dr. Epstein at the annual meeting of the American Urological Association (AUA) in New Orleans a few weeks ago. We had not reported on this before because we wanted readers to have access to a version of this information that clearly presented the new system in Dr. Epstein’s own words.

Basically, the new grading system will be used to divide localized prostate cancers into a series of five histopathological groups, numbered simply from 1 to 5. However, for at least a while, in Dr. Epstein’s words,

The new grading system, recently accepted by the [World Health Organization] will be used in conjunction with the Gleason system until it becomes widely accepted and practiced.

The basic premise is that all prostate cancers can be subdivided, histopathologically, into five categories described in the table below:

Picture2

Now this is all a little technical, and it is going to take a while for us all to get used to, but there are three important points that we will all need to understand as we start to adapt to this new prostate cancer grading system:

  • It gets rid of  all the inherent complications of the existence of the Gleason scores from 1 + 1 = 2 to 3 + 2 = 5, none of which are currently reported any more anyway, but whose existence prejudices everyone’s thinking because the lowest reported Gleason score today is 3 + 3 = 6.
  • It deals effectively with the fact that Gleason 3 + 4 = 7 really is a different form of histology compared to Gleason 4 + 3 = 7.
  • It differentiates between cancers that are more like Gleason 4 + 4 = 8 and those that are more like Gleason 9s and 10s.

Thus, we will be able to return to a simpler system in which Grade 1 is the least aggressive grade and Grade 5 is the most aggressive grade and stop worrying about all the differences introduced by primary, secondary and tertiary patterns of cancer.

As Dr. Epstein explains in his article, this new system has been validated through a careful analysis of whole-gland pathological specimens and biopsy cores from tens of thousands of specimens at some of the world’s leading prostate cancer treatment and pathology centers. In addition, the five new grade groups are clearly different in their outcomes (as assessed by risk for 5-year biochemical recurrence) when men with these grades are treated by radical prostatectomy. Furthermore, cancers that are defined as Grade 1 according to the new system have no potential whatsoever for transition to metastatic disease.

Professional pathologists, just like the rest of us, will need time to learn how to read prostate cancer pathology and biopsy slides and grade cancers using this new system, which is why it is a good idea for us to accept the idea of a significant period of transition from the old Gleason system to the new “Epstein” grades.

The “New” Prostate Cancer InfoLink will, obviously, start to slowly adopt and adapt to this new prostate cancer grading system. Whenever possible, we will provide both old and new grading information for a while. We agree with Dr. Epstein — and all of the others who have helped to put this new grading system in place — that this new system will prove to be a significant advance, from a clinical perspective, over the various evolutions of the Gleason scoring system (which was last revised, yet again, in 2014). However, we also recognize that this new system is going to require a complete overhaul of prognostic systems like the Kattan nomograms and the Partin tables that have been so helpful in offering patients prognostic information that is useful in making decisions about treatment options.

We all going to need a while to adapt to this new system. We suspect that it will become known as the Epstein grading system over time — and we also suspect that Dr. Epstein will do his very best to avoid the use of such terminology at all costs!

27 Responses

  1. I’d be interested to know what Gleason score (say Gleason 6) corresponds to what new number. (For example, Gleason 6 = 2 on the above chart.)

  2. Walt:

    There is no precise correlation between the old Gleason scores and the new Epstein grades. This is all based on the histology, not on the old Gleason scores. Thus, the current Gleason 3 + 3 = 6 is not an exact match for the new Epstein stage 1 and the old Gleason scores of 9 and 10 are not an exact match for the new Epstein stage 5. However, there are similarities, and as an approximation one can think about the new system more like this:

    — Epstein grade 1 is similar to any Gleason score of 6 or less
    — Epstein grade 2 is similar to a Gleason score of 3 + 4 = 7
    — Epstein grade 3 is similar to a Gleason score of 4 + 3 = 7
    — Epstein grade 4 is similar to a Gleason score of 4 + 4 = 8
    — Epstein grade 5 is similar to a Gleason score of 9 or 10

    But … Dr. Epstein would want me to point out carefully that “similar” is a long, long way from being “the same”.

    As an example, Dr. Martin Luther King and I have many similarities: we are both human; we both had a good education; we both were born in the first half of the 20th Century; most people would probably consider both of us the be “liberal” in our political beliefs; etc., etc. But … our differences probably far outweigh our similarities, starting with the fact that he was a heck of a lot more determined to achieve his goals than I have been to achieve many of mine over the years!

  3. Not sure what any of this has to do with treatment recommendations based on so called stage of the disease at pathology or biopsy. What will be done differently based on this new definition? It looks like the major beneficiaries will be those for whom no treatment is necessary , not for those of us with Gleason 9, pT3b disease.

  4. I guess what remains to be seen is how the new categories will be subsumed into revised risk stratification schemes. Epstein has written before that even a minor presence of tertiary Gleason pattern 5 significantly raises the risk of men with predominantly Gleason 7s to a high-risk category, but seems to have dropped that idea here. I also wonder how long it will take to catch on.

  5. Dear Bob:

    This new system is not going to make a lot of difference to treatment recommendations for men diagnosed with very high-risk prostate cancer like yours. We have known for years that men like you need aggressive treatment as early as possible to minimize risk for progression and prostate cancer-related death.

    On the other hand, over time, these distinctions are likely to make a very great deal of difference to the treatment recommendations for men who have been diagnosed, historically, with Gleason score of 3 + 4 = 7 and lower and a clinical stage of cT1 or CT2a, and this is the vast majority of the 230,000 men who get diagnosed with prostate cancer in America each year.

  6. Dear Allen:

    The new histological characterization actually ensures (for example) that any presence of what is currently classified as tertiary Gleason pattern 5 would raise the risk classification of the patient (say from Epstein 2 or Epstein 3 to Epstein 4 or Epstein 5, depending on how much of the tertiary pattern tissue was evident).

    With respect to how long it will take this new system to catch on, my guess would be that it will be in widespread use at major centers in the western world within 3 years because this will be driven by a relatively small number of specialized uropathologists who act as consultants to all of the major laboratory testing services quite apart from what they do at their own laboratories. You need to appreciate that within the uropathology community this concept has been under discussion for several years now, so it is not exactly a new idea. It just took time to sort out all of the details.

  7. Do you know if Dr. Bostwick of Bostwick Laboratories has bought in to this? He and Dr. Epstein disagreed when the Gleason definitions were changed several years ago. It would be good for everyone if these two highly respected pathologists were in agreement on this.

    Jim

  8. Jim:

    It is my understanding that Dr. Bostwick is no longer involved in the day to day running of Bostwick Laboratories because of his other business interests.

  9. Another question I have is not related to the demographics among pathologists, but rather how this will effect existing clinical trials and treatment guidelines. It will almost be necessary for both grading systems to be used until a prospective trial comparing the two grading systems can be completed.

    I have always admired Jon Epstein’s work in prostate cancer pathology and research. Other than Donald Gleason himself, perhaps, he has been the best at this in history. The few pathologists I know would not argue that point.

  10. Tony:

    (1) I have no doubt that the two systems will need to be used side by side for a considerable period of time for all sorts of reasons (including the clinical trials issue). And I am sure that the leading pathologists have already planned to put some form of comparative monitoring system in place to check on the relative accuracy of the Gleason system and the new system in practical, day-to-day use over the next several years.

    (2) I think it is also important to recognize that although Dr. Epstein has certainly been a key “driver” behind the adoption of the new grading system, he is most certainly not the only person behind this. The impression I have is that the majority of the leading uropathologists are fully on board with this. I’d be interested, for example, to know what the key SWOG-associated uropathologists were saying about the implementation of this switch over time.

  11. Given that this is very new, I don’t have the answer from our committee pathologists. Given, however, that Epstein is regularly tagged in the past in our trials I would expect favorable acceptance.

  12. Tony:

    At least some of the leading SWOG uropathologists must have been involved in the discussions that have been ongoing about all of this. I have a hard time thinking that this is “new” to them. They have to have known this was coming. It’s not as though Epstein did this all on his own at all.

  13. Agreed. It’s more new to me. Interestingly, I attended a da Vinci display with a Cleveland Clinic surgeon yesterday at the University of Nevada, Las Vegas and found it astonishing that the Las Vegas valley has but one true “uropathologist” after a couple retired or moved out of the area. We agreed that that one pathologist is hardly enough to handle a community as large as the southern Nevada territory and most biopsies here are defined by pathologists overwhelmed with all forms of tissue samples from all diseases. Discouraging as that sounds, bring in this discussion and it hardly seems possible that 3 to 5 years adopting a new system is likely here. It also makes me even more inclined to suggest to the newly diagnosed here to get the slides sent to where well-known pathologists are. Understanding that Bostwick is now doing things other than prostate cancer slide readings, I now have to believe that these readings are best at major institutions with strong background in prostate cancer care.

  14. Tony:

    Let me be clear that just because Dr. Bostwick is no longer personally reading prostate slides does not mean that the company has been left in the hands of people who are not expert in uropathology. That is the entire company’s primary business focus.

  15. Thanks for posting this information! I’m going in for prostate cancer testing, so knowing the new cancer grading system will help give me a better understanding of what’s going on. It’s interesting that prostate cancer can be subdivided into five categories. I didn’t realize just how complicated the Gleason scoring system was. It seems like doing all of that math would make it a bit more difficult to accurately grade cancer. Hopefully, this new system makes it much easier for doctors to grade prostate cancer in patients.

  16. “Furthermore, cancers that are defined as Grade 1 according to the new system have no potential whatsoever for transition to metastatic disease.”

    So, I read the link in Epstein’s article that references this. In the article Epstein references a study that showed in over something like 14,000 cases the study did not find one case of <=G6 spreading to lymph nodes. Epstein then states based on this study <=G6 does not have metastatic potential. Excuse me if I am wrong, but is the lymphatic system the only way cancer can spread? What about through the blood stream? How can such a generalized statement be made solely based on a study that did not find it in the lymph nodes only?

    If this is true, then how do you explain the BCR free rate of 95% for the <=G6? Unless this data includes positive margin patients, shouldn't it be 100% if there is not metastatic potential?

  17. Chris:

    (1) Numerous studies have now confirmed the data originally published by Epstein and his colleagues that cancers with a Gleason score of 3 + 3 = 6 or lower have no potential to metastasize (to the lymph nodes or anywhere else). And no, not every prostate cancer metastasizes via the lymph nodes … but the vast majority do.

    (2) You need to appreciate that just because a cancer cell travels outside the prostate (regardless of how and/or to where) doesn’t mean that it can start to grow again elsewhere if it doesn’t have the right biological characteristics. Metastasis requires a very particular series of complex events to take place in the right cellular microenvironment.

    (3) The fact that the BCR-free rate is said to be 95% for men with a Gleason score of 6 is not based on central pathological analysis of all the men said to be Gleason 6 and it does not take account of cancers for which there were positive margins or other forms of extracapsular extension that weren’t completely eliminated at surgery. What Epstein’s group and other groups did was go back and look carefully at all of the pathological specimens that were said to be Gleason 6 and which did not have any indication of cancer left behind after surgery. What they then found was that there were a small percentage of those Gleason 6 or lower cancers that were not, in fact Gleason 6 at all, but on closer examination were found to be of higher Gleason score or to have a higher tertiary Gleason pattern. Those cancers were capable of metastasis; the “real” Gleason 6 and lower Gleason score cancers were not.

    (4) You also need to recognize that just because a patient has a biochemical recurrence of a Gleason 6 cancer because of a positive margin still doesn’t mean that the cancer has the potential to metastasize. The patient’s PSA can rise but the cancer may be confined to a relatively small area of the prostate bed.

    The 95% BCR-free rate among patients with Gleason scores of 6 or less basically allows for less than perfect surgery and less than perfect pathological assessment of the post-surgical specimens. The new system should actually help to minimize risk for the latter.

  18. OK. Then … based on all this, I was organ-confined Gleason 6 post-op from Dr. Partin in 2008. Epstein, in an e-mail this morning to me responding to me asking him how I could get my post-op slides re-graded in the new system, told me that there was no need: that I was Grade 1 in the new scale and nothing would change if I was organ-confined Gleason 6 in 2008 graded by them.

    I suspected that when Epstein produced the 2005 ISUP guidelines that narrowed down the Gleason 6 classification he was driving towards a way to determine what men with prostate cancer would not progress with metastatic cancer and to explain why the small number of men with Gleason 6 did. By narrowing down the criteria for becoming a Gleason 6, you would be able to tell who was effectively cured — with the understanding that some pathologies were upgraded due to other reasons as well.

    Based on all of this I can stop PSA testing since there is no potential for my prostate cancer to have metastasized. My last PSA was done a year ago and was < 0.03 which was the 6-year mark. … :) Maybe I'll do one more.

  19. Has anyone checked with the clinicians to see if they want this? None of the physicians in our urology group have heard of this and don’t know what to do with it. We are all very familiar and happy with basing our treatment decisions on Gleason grading. Is Epstein just trying to secure his legacy and name with his own grading system and push Gleason’s 40 years of tried and true grading to the curb?

  20. Dear Dr. Wagner:

    I am quite sure that leading members of the pathology community will have discussed this extensively with many of their urology colleagues. There seems to be a clear consensus in the uropathology community that the proposed new system is more accurate than the Gleason system that has been revised multiple times in recent years. And it has also been made very clear that everyone understands the need for a long period for full adoption of the new system.

    Change can be difficult — but sometimes change is also necessary to offer patients the highest quality of care.

    With respect to Dr. Epstein, I think it is highly unlikely that the question of a “personal legacy” crossed his mind — any more than it did in the case of Dr. Gleason some 40 or so years ago now.

  21. As I mentioned above, what really matters is how categories are used for risk stratification. A new study demonstrates that Gleason 3 + 5 and Gleason 5 + 3 have significantly worse survival prognosis than Gleason 4 + 4. These are all lumped together in Epstein’s new Grade 4. It doesn’t seem to serve any risk stratification purpose to do so.

  22. Dear Allen:

    The “significantly worse prognosis” you refer to in this study is based on a retrospective analysis of data from just 26 deaths among 462 patients over a period of 14 years. While one may be able to make a statistical argument based on these data, I have my considerable doubts about any real clinical significance.

  23. The median follow up was 7.6 years. In that time, the all-cause mortality was 26%, and mortality that was specifically attributed to prostate cancer was 6%. These numbers are similar to those reported by another recent study of LDRBT boost therapy to men with Gleason pattern 5: Liss et al. reported 5-year CSS of 93%. The fact that men were 2.8 times as likely to die of prostate cancer if they had Gleason 3 + 5or 5 + 3 seems both meaningful and statistically significant to me, even with this small base of deaths.

    I do agree with you that it was a retrospective study, and that outcomes may change with longer follow up. Gleason pattern 5 is particularly virulent, and is known to be more radioresistant, more chemoresistant, more hormone-resistant, and to have a very different biology (putting out less PSA, for example) than other Gleason patterns. Because of that, I would expect the gaps to widen with prospective tracking and longer follow up.

    In risk stratification, we seek groups that are homogeneous within the group and heterogeneous between groups. This study identified one of the three “new Grade 4” definitions that was, instead, heterogeneous from the other two in its group.

    The real point is that if we were to use Epstein’s new grading system, we would lose the ability to make such distinctions and identify subgroups that may respond differently to any of a number of treatments according to their GS x+y status. I sympathize with his goal of wanting to reduce the cancer panic attributable to patient’s hearing higher numbers, but would very much regret the loss of potentially valuable information by lumping those groups together.

  24. Dear Allen:

    Why do you think that the patients that are Gleason 3 + 5 or 5 + 3 in the studies you are referring to are necessarily ISUP grade 4? As I understand the ISUP system, if the tissue samples contained any amount of tissue that “lacks gland formation — or has glands with necrosis — with or without poorly formned, fused, and/or cribriform glands” then they would be ISUP grade 5.

    You seem to be under the impression that any cancer that has a Gleason sum of 8 is necessarily ISUP grade 4. That is absolutely not my understanding at all.

  25. Dear Allen:

    Why do you think that the patients that are Gleason 3 + 5 or 5 + 3 in the studies you are referring to are necessarily ISUP grade 4? As I understand the ISUP system, if the tissue samples contained any amount of tissue that “lacks gland formation — or has glands with necrosis — with or without poorly formned, fused, and/or cribriform glands” then they would be ISUP grade 5.

    You seem to be under the impression that any cancer that has a Gleason sum of 8 is necessarily ISUP grade 4. That is absolutely not my understanding at all.

  26. There are three patterns identified in his ISUP grading system:

    — “well-formed glands” – consistent with Gleason pattern 3 (and the sole component of ISUP Grade 1)
    — “poorly formed, fused, and/or cribriform glands” — consistent with Gleason pattern 4 (and it MAY be the sole component of ISUP Grade 4)
    — “lacks gland formation” – consistent with Gleason pattern 5 (and it MAY be the sole component of ISUP Grade 5)

    In his new system he uses three definitions within ISUP Grade 4:

    (1)”ONLY poorly formed, fused, and/or cribriform glands”, i.e., Gleason 4 + 4

    (2)”Predominantly well-formed glands” (i.e., Gleason pattern 3) “WITH lesser component lacking gland formation” (i.e., Gleason pattern 5) — so this equates to Gleason 3 + 5

    (3)”Predominantly lacking gland formation” (i.e., Gleason pattern 5) “AND lesser component of well-formed glands” (i.e., Gleason pattern 3) — so this equates to Gleason 5 + 3 (he allows tertiary 4 as a minor component)

    Gleason pattern 5 — “lacks gland formation” — is included in BOTH ISUP Grades 4 AND 5.

  27. Dear Allen:

    I understand your argument, but neither you nor I are pathologists — let alone specialized uropathologists. It is really rather pointless you and I discussing this in detail. Now if you would care to take this up directly with Dr. Epstein …

    :O)

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: