What should a “castrate” level of serum testosterone actually be?


For many years now, there have been subsets of specialists and patients who have argued that the “standard” serum testosterone level of 50 ng/dl, used by regulatory authorities and others as a measure indicating a castrate level of serum testosterone, is too high. A new paper by Bryant et al. in the “Red Journal” would seem to clearly confirm this belief.

Perhaps it is time for the Prostate Cancer Working Group, the U.S. Food and Drug Administration, and others to take note and make some changes.

Bryant et al. looked at long-term data from a cohort of 764 US veterans, all of whom were initially diagnosed with intermediate- or high-risk, localized prostate cancer, and all of whom were treated with a combination of androgen deprivation therapy (ADT) and definitive radiation therapy between 2000 and 2015.

They divided the patients into two groups based on their nadir serum T levels post-treatment, while the patients were on continuous LHRH agonist therapy:

  • Group A: Patients whose nadir serum T level was < 20 ng/dl
  • Group B: Patients whose nadir serum T level was 20 to 49 ng/dl

They then looked at a spectrum of patient outcomes over time.

Here is what they found:

  • Compared to patients in Group A, patients in Group B had
    • A higher 3-month post-radiotherapy serum T level
    • A higher 2-year post-radiotherapy serum T level
  • 10-year biochemical recurrence rates were
    • 18.3 percent for men in Group A, as compared to
    • 28.1 percent for men in Group B
  • Rates for onset of metastatic disease were
    • 7.8 percent for men in Group A, compared to
    • 12.9 percent for men in Group B
  • Rates for prostate cancer-specific mortality were also higher in Group B than in Group A (although the difference was not statistically significant).

Bryant et al. conclude that:

Additional serum testosterone suppression below 50 ng/dl was associated with improved PSA responses and lower rates of biochemical recurrence and metastasis in this cohort of localized [prostate cancer] patients.

With the availability of multiple new types of androgen suppressing agents over the past 30 years, and with an increasingly compelling amount of evidence to suggest that men with higher-risk forms of prostate cancer do better on ADT (with or without radiation therapy) when their PSA levels are lowered to significantly lower than 50 ng/dl, it is time for at least two things to occur:

  • It should become standard practice for physicians to measure the serum testosterone levels of all men on ADT to ensure that they actually have a castrate level of serum T.
  • Appropriate organizations need to reach consensus on what castrate levels of serum T — and arguably of serum dihydrotestosterone or DHT too — really should be.

The available data would currently suggest that a truely castate level of serum T should be at least < 35 ng/dl and arguably (on the basis of this paper by Bryant et al.) as low as <20 ng/dl.

As far as we are aware, no one has ever set out to determine what a castrate level of serum DHT should really be, but one could certainly argue that it should be no higher than the serum DHT level of men with a serum T level of < 20 ng/dl.

Our understanding is that the current standard castrate level of serum T was originally set at < 50 ng/dl back in the 1970s — but it might have been set even earlier than that. Until 1995 there was no accurate mechanism to measure serum T levels lower than 50 ng/dl. We have now had better methods for 23 years. It is time to take advantage of such techniques and to ensure that they are applied in clinical practice.

18 Responses

  1. “Compared to patients in Group B, patients in Group A had

    — A higher 3-month post-radiotherapy PSA level
    — A higher 2-year post-radiotherapy PSA level”

    Is this the right way around?? … of course, I could be misreading but am confused.

    Happy New Year Mike – may 2019 be happy and healthy!

  2. Sorry, but I don’t think this comment perfectly drawn the most useful conclusion of the study. I believe that important point is the overall long-term outcome from the nadir value of the treatment. So if some patients get a serum T nadir below 20 ng/dl, fair enough, but for those who have single-agent ADT in future (if that still exists), and their nadir serum T level remains above 20 ng/dl. The bad perspective motivates an additional effort to intensify single-agent ADT with second generation AR inhibitors like abiraterone or enzalutamide, or with chemotherapy like docetaxel. There is all to win with intensifying treatment to get a further lowered nadir serum T level and it is relevant to go for that the additional reduction of serum T might increase progression-free survival and reduce death of prostate cancer compared to action later in the clinical course where the patient develops PSA recurrence and then syart with these new effective drugs.

    Also, we have randomized trials supporting the combinations at the start of ADT. So the point is not only to reduce testosterone, and the level of testosterone might not be the only reason why some patients have higher PSA nadirs than other patients. A patient might achieve the low testoserone levels and nevertheless not have the best low post-treatment nadir PSA.

  3. Bravo! Hooray! Amen! Finally!

    Thanks very much for posting this!

    There is a minor flaw at:

    “They divided the patients into two groups based on their nadir PSA levels post-treatment, while the patients were on continuous LHRH agonist therapy:

    Group A: Patients whose nadir PSA level was < 20 ng/dl
    Group B: Patients whose nadir PSA level was 20 to 49 ng/dl"

    It's easy to see what you meant, but PSA should obviously be replaced by testosterone.

  4. The Ideal Castrate DHT Level for Prostate Cancer Patients on ADT

    Sitemaster, you wrote above: “As far as we are aware, no one has ever set out to determine what a castrate level of serum DHT should really be, but one could certainly argue that it should be no higher than the serum DHT level of men with a serum T level of < 20 ng/dl."

    Good point!

    I am not aware of guiding research either. However, for years Dr. Charles "Snuffy" Myers advised that patients should achieve a serum DHT level of 5 or < 5 ng/dl (don't recall which), while also achieving a serum T level of < 20 ng/dl. This was rather a big deal for him personally, as he was one of those patients who had a rather high serum DHT level despite being on effective Lupron. My recollection is that his DHT for his aggressive case of metastatic prostate cancer was in the 50s, which, as DHT is far more potent as a fuel for prostate cancer than testosterone, would have absolutely wrecked his androgen blockade program if not corrected (which he did).

    This may be one of the reasons why he has considered a 5-alpha-reductase drug (Proscar/finasteride or Avodart/dutasteride) as an important element of ADT therapy, because both drugs greatly reduce the conversion of any remaining testosterone to DHT. Dr. Myers, before his recent retirement, was one of the few medical oncologists in the US with a large practice in a non-academic setting dedicated to prostate cancer. His practice involved extensive and rigorous monitoring with many patients having advanced cases, and this large throughput of data, electronically retrievable, enabled him to see patterns, such as the effects on success of different levels of DHT.

    It would be a lot better to have solid, published, trial-based research, but in the meantime the views of insightful physicians deserves weight.

  5. It would be very unfortunate, in my opinion, if the headline were heeded by the powers that be, with the change consisting only of redefining a word to mean something other than (or even just a subset of) what it has meant for a long time.

    That approach leads to confusion for everyone who ever looks back in the literature and must somehow know that the word as used in 2040 meant something different in 2020 and something different yet again in 2010. That’s how we ended up with the term “undetectable” meaning “PSA <0.1". Let "undetectable" go back to meaning undetectable (in the context of year and assay), and let a measured detectable value of "PSA =0.08" never forthwith be considered "undetectable".

    If the Red Journal results are replicated, then a new recommendation would seem warranted:

    • If a new recommendation is made to bring T to below 20 ng/dl, then the new recommendation should be to bring T to below 20 ng/dl.

    • If a more nuanced and flexible recommendation is made to bring T to the lowest level consistent with the patient's QoL and pocketbook in the milieu of their country's current pharmamentarium, then call it "Recommendation PCa2019Q17b" or some such.

    But please don't encourage anyone to redefine meanings

    In my strong and indignant opinion, the term "castrate" should never have been repurposed to mean "T at or below a certain level".

  6. Why isn’t Avodart considered an important supplement to LHRH agonist or antagonist therapy, since it’s a 5-AR inhibitor?

  7. Re-read the corrected version. Mea culpa!

  8. Dear Finn:

    I think you may have been confused by an error in the original post that has now been correctect. I suggest you re-read the post. This is not about PSA levels at all. It is all about serumn testosterone and serum dihydrotestosterone levels. If these are sufficiently low, then clearly the patients have better outcomes over time.

  9. Jim:

    Yes. This has been corrected. Mea culpa

  10. Dear Bob:

    There are several problems related to why neither finateride/Proscar nor dutasteride/Avodart are used routinely in the treatment of prostate cancer in conjunction with LHRH agonists and antagonists:

    — No large, randomized clinical trial has every been carried out that showed a significant clinical benefit (probably because to be effective it should only be used in men who have an elevated serum DHT level on an LHRH agonist alone)

    — All 5-AR inhibitors can have a low level of significant side effects over and above those of the LHRH agonists and so if there is no proven clinical benefit, why would one use them?

    — There are no guidelines stating that use of 5-AR inhibitors is appropriate in the treatment of progressive or advanced prostate cancer

    What is almost certainly going to be needed is a randomized trial that clearly demonstrates that, in men with an elevated serum T level and an elevated serum DHT level after (say) 6 months of treatment with an LHRH agonist, adding a 5-AR inhibitor can lower serum T and serum DHT levels to an appropriate, pre-defined level. Ideally such patients should then be followed for long enough to see if this impacts time to metastasis or overall survival. However, what is rather less clear is who might be willing to pay for such a trial.

  11. Sitemaster:

    Thanks. I guess another point is if T is well below 20 without a 5-ARI there’s no need to stop DHT from turning into T.

    I was on intermittent triple blockade for 3 years but was advised (based on recommendation of my RO, who is an advocate of Snuffy Meyers) not to drop Avodart and Casodex when I went on Lupron holiday. Now I’m on Trelstar only … waiting to see if PSA continues to decline.

  12. Dear Bob:

    The precise subtleties of how to minimize risk of prostate cancer progression once it becomes micrometastatic are more complex than most people realize, and the data that support specific perspectives probably apply to potentially identifiable subsets of people rather than to all prostate cancer patients.

    If I was diagnosed with metastatic prostate cancer tomorrow, I would be politely advising my MED/ONC that I wanted him (or her) to be monitoring my serum T level, my serum DHT level, my PSA level, and my PAP level until we at least had a clear understanding of what they were and whether they were stable on a specific form of treatment. However, … I am also highly conscious that even if all four of them were low and stable, progression could occur at any time as a consequence of evolutionary progression based on further metastasis.

  13. Dear Paul:

    The problem with the word “castrate” in the context of castrate-resistant prostate cancer is that it was never accurately defined to begin with, for the simple reason that the available test couldn’t measure it accurately enough. As a consequence, I have never likes it anyway.

  14. Bob:

    I have known men with morphed disease at an advanced stage to have very high DHT levels despite low T … so at some point the 10% rule may not hold for all. Purely anecdotal, but I can attest to it.

  15. Sitemaster:

    Happy New Year and today is my 75th birthday!

    I’ve been very sensitive to ADT3. My T has hovered from 3 to 7. I’ve always had T, DHT, and PAP measured along with PSA. I don’t understand why anyone wouldn’t.

    I agree that no one will ever do a trial on 5-ARI meds. Dr. Myers — as we know — did his own mini trial and recommended staying on them.
    I’m doing my own mini trial using an LHRH agonist only to see what happens to T and PSA.

  16. Hi again Mike,

    There is still a flaw in your review; it reads:

    “Here is what they found:

    Compared to patients in Group B, patients in Group A had
    A higher 3-month post-radiotherapy serum T level
    A higher 2-year post-radiotherapy serum T level”

    The B and A should be reversed. The patients in Group A did better.

  17. Whether to revise the language for a “castrate” level of testosterone: Do it!

    Hi PaulC. I’m responding to your comment of December 31, 2018 at 6:49 pm regarding a couple of points.

    I strongly favor use of a level of around < 20 ng/dl as the threshold for doctors who are trying to achieve effective ADT, but, because a level of < 50 ng/dl has been used for so long and almost universally in published research, I would like to see wording like "revised castrate level" to describe the new threshold and something like "traditional castrate level" to describe the old < 50 ng/dl level. (Indeed, "traditional" is a much more polite term than "obsolete, misleading and risky castrate level", which hits the nail right on the head but would no doubt be uncomfortable and upsetting to conventional wisdom).

    Such distinctive words — "revised" and "traditional" — would call attention to the change. That is needed as there is enormous inertia in medical practice and a shake up is needed in thinking about testosterone and ADT. While people who pay attention would still be fairly aware of the change to a lower threshold and its timing yearwise without changed wording, that group is likely quite small, even among physicians and many researchers, with the vast majority of patients being almost clueless. I'm going to start using "revised castrate level" and "traditional castrate level" myself, hoping it will foster awareness.

    Changed wording here has a prominent precedent. I am aware of one very salient terminology change regarding testosterone and ADT that occurred since I became a prostate cancer patient in late 1999. Back then, and for a number of years thereafter, the term for what later rather abruptly changed to "castrate level" or "castrate resistant" (as in the commonly used acronym "CRPC") was "hormone resistant", with the associated acronym "HRPC". I was rather surprised at how rapidly terminology changed in published research.

    Also, the doctors I follow closely did not find it either hard or expensive to bring testosterone down to the lower level for most of the patients who did not initially achieve that level (a fortunate thing, because abiraterone acetate and Xtandi were years away from discovery). These doctors frequently were consulted by patients who were not doing well with other doctors, and a primary corrective tactic was (is) to ensure that the ADT was being correctly administered, such as: into the butt muscle and not fat for Lupron; prevention of the coil slipping out for Zoladex; proper storage of the medication; and proper mixing of the medication. As a patient, I always watched to make sure the nurse or doctor administering my Lupron shot was mixing it correctly. A second corrective tactic was (is) to monitor testosterone sufficiently often to determine that the patient does not clear the depot/coil medication unusually early; if clearance more rapidly than normal is detected, then the dosing interval is shortened appropriately, such as giving a 3-month shot at a 2.5-month interval (my layman's guess as an example at the adjustment that might be used).

    (The expert doctors were also monitoring DHT — and some other meaningful biomarkers — and adjusting therapy to hit a low threshold, but that is a different story.)

  18. Tx. Duly corrected.

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: