Updated overall survival after abiraterone acetate in mCRPC


According to a new article, to be published on line today in The Lancet, abiraterone acetate + prednisone provides a median overall survival benefit of 4.6 months compared to a placebo in men with metastatic, castration-resistant prostate cancer (mCRPC) who have received prior treatment with docetaxel-based chemotherapy.

As yet, the actual article is not available on The Lancet‘s web site, but a summary of the report has been published on the MedPage Today web site.

According to the summary report, this is the “final” analysis of the original trial initiated by Cougar Pharmaceuticals. It shows that over a median follow-up period of 20.2 months, median overall survival was

  • 15.8 months for men randomized to treatment with abiraterone acetate + prednisone
  • 11.2 months for men randomized to a placebo + prednisone

Additional data provided show that:

  • The final analysis was carried out before any patients were crossed over from placebo to abiraterone and after 775/797 prespecified death events.
  • The finding of an overall survival benefit was consistent across all the protocol-specific patient subgroups.
  • Median time to biochemical progression (based on a rising PSA value) was
    • :8.5 months in men randomized to abiraterone acetate
    • 6.6 months in men randomized to the placebo arm
  • Median time to radiologic progression was
    • 5.6 months in men randomized to abiraterone acetate
    • 3.6 months in men randomized to the placebo arm
  • The most common grade 3-4 adverse events were fatigue, anemia, back pain, and bone pain (and these were all observed at similar frequencies in both arms of the study)
  • Mineralocorticoid-related adverse events occurred more often in men in the abiraterone arm of the trial; so did fluid retention, edema, hypokalemia, and hypertension.

These new data do show that abiraterone acetate offers a somewhat longer overall survival benefit than the data presented when the trial was originally stopped in 2010. The data report suggest that the “final” overall survival benefit of abiraterone acetate is closer to — but still shorter than — the overall survival benefit offered by data from the first analysis of the enzalutamide trial in a comparable group of patients with mCRPC. The only way we will know if one of these products is actually “better” than the other is if someone decides to conduct a head to head trial … and we aren’t expecting to see such a trial in men with mCRPC any time soon, if ever.

15 Responses

  1. This indicates that radiographic progression was 3 months earlier than PSA progression. Is that right? Is this common to other trials in similar populations? Is PSA so unreliable that a rise is seen only months after new metastases are already seen on imaging?

  2. The key questions for patients are now:

    (1) Should abiraterone or enzalutamide be used first?

    (2) Will the overall responses to the two drugs be the same, regardless of which one is used first and which second?

    (3) Why not use both at the same time?

  3. I found a paper presented by Scher et al. at the recent ESMO meeting (889PD Association of baseline corticosteroid with outcomes in a multivariate analysis of the Phase 3 AFFIRM study of enzalutamide [ENZA], an androgen receptor signaling inhibitor) very interesting. I would love to hear your take on this.

  4. Dear Tarhoosier:

    First, there has been only one other large trial of this type The pivotal trial of carbazitaxel), and I am not sure that the researchers measured time to radiographic progression in that trial, so I simply can’t tell you whether the effect described is “common to other trials in similar populations.”

    Second, what these data actually seem to be telling us is that — at least in this group of patients who have progressed after androgen deprivation and chemotherapy — neither biochemical failure due to a rising PSA nor the appearance of new or enlarged metastases is necessarily a good surrogate marker for overall survival.

    I think that — at this time — the only thing we can conclude with certainty from this study is that real overall survival is the only trial endpoint that can tell us with accuracy what the effect of a specific drug on overall survival may be in men with mCRPC who have already been given chemotherapy..

  5. >(3) Why not use both at the same time?

    Another ESMO paper presented yesterday, by Johann de Bono, and titled “Integrating novel endocrine therapies: sequential or concomitant treatment” is precisely on this topic.

    According to the abstract of this paper, “… these two drugs in combination may be superior to either drug alone. We have hypothesized that progressive disease on abiraterone may occur secondary to glucocorticoid-induced activation of mutated AR. We … found that prednisolone plasma levels in CRPC patients were sufficiently high to activate mutant AR. Mineralocorticoid receptor antagonists, such as spironoloactone and eplerenone that are used to treat side-effects related to mineralocorticoid excess, also bound to and activated signaling through both wild-type and mutant AR. Activation of mutant AR was inhibited by MDV3100, bicalutamide or greater concentrations of abiraterone. Moreover, higher androgenic steroid concentrations resulted in decreased enzalutamide antitumour activity. Together, our findings provide a strong rationale for the clinical evaluation of combined CYP17A1 inhibition and AR antagonism by enzalutamide.”

  6. The difference in overall survival benefit between Zytiga (abiraterone) of 4.6 months and Xtandi (enzalutamide) of 4.8 months is only 0.2 months (i.e., days) which is too close to call in a clinical trial setting.

    There is one important difference between the abiraterone and enzalutamide trials:

    — Abiraterone + prednisone was compared to placebo + prednisone.

    — enzalutamide was compared to a true placebo.

    This means that in the abiraterone trial the overall survival is compared to treatment with prednisone. This drug has anticancer properties of its own and is used in combination therapy with docetaxel and mitoxantrone as well as abiraterone.

    The abiraterone trial results showed that prednisone alone induced a 5% response rate in lowering PSA levels.

  7. Dear Gerry:

    There is a second issue that is important. Like the abiraterone trial, the enzalutamide trial was stopped early. We do not yet know what the “final” results of the enzalutamide trial might show. At the time of the publication of initial analysis of the abiraterone trial, the overall survival benefit of abiraterone was 3.9 months (as opposed to 4.6 months in the above “final” analysis). It is perfectly reasonable to think that the “final” analysis of the enzalutamide trial might also show longer survival (i.e., something in excess of 5.5 months) in men with mCRPC who have received at least one cycle of docetaxel-based chemotherapy. We are just going to have to wait and see.

  8. Dear Gerry:

    There is a second issue that is important. Like the abiraterone trial, the enzalutamide trial was stopped early. We do not yet know what the “final” results of the enzalutamide trial might show. At the time of the publication of initial analysis of the abiraterone trial, the overall survival benefit of abiraterone was 3.9 months (as opposed to 4.6 months in the above “final” analysis). It is perfectly reasonable to think that the “final” analysis of the enzalutamide trial might also show longer survival (i.e., something in excess of 5.5 months) in men with mCRPC who have received at least one cycle of docetaxel-based chemotherapy. We are just going to have to wait and see.

  9. Dear Summer:

    The paper you refer to basically suggests that men who are taking corticosteroids (e.g., prednisone or prednisolone or dexamethasone) at the time that they started taking enzalutamide in the AFFIRM trial did less well (i.e., had a shorter survival) than men who were not taking corticosteroids.

    The is issue here is: which is the chicken and which is the egg? In other words, did these men do worse because they were on the corticosteroid therapy or does the fact that they were on corticosteroid therapy suggest that they had more advanced disease than the men who were not on corticosteroid therapy (in which case it would be reasonable to expect them to do less well than the men with less advanced disease who did not need corticosteroid therapy)? I don’t think there is necessarily a good answer to this question yet. It is certainly not addressed in the abstract of the paper, so we may need to see the full publication (unless you want to contact Dr. Scher and ask him what he thinks).

  10. Dear Summer:

    The paper you refer to basically suggests that men who are taking corticosteroids (e.g., prednisone or prednisolone or dexamethasone) at the time that they started taking enzalutamide in the AFFIRM trial did less well (i.e., had a shorter survival) than men who were not taking corticosteroids.

    The is issue here is: which is the chicken and which is the egg? In other words, did these men do worse because they were on the corticosteroid therapy or does the fact that they were on corticosteroid therapy suggest that they had more advanced disease than the men who were not on corticosteroid therapy (in which case it would be reasonable to expect them to do less well than the men with less advanced disease who did not need corticosteroid therapy)? I don’t think there is necessarily a good answer to this question yet. It is certainly not addressed in the abstract of the paper, so we may need to see the full publication (unless you want to contact Dr. Scher and ask him what he thinks).

  11. If anyone can find a copy or a video of the actual presentation by Dr. de Bono mentioned above, that would be interesting to see.

  12. >We do not yet know what the “final” results of the enzalutamide trial might show.

    Actually, the updated results was published last week as part of the FDA review. The medians dropped slightly.

    The updated medians are 17.8 vs 13.3. The new HR is 0.62 (0.52, 0.73).

  13. >I don’t think there is necessarily a good answer to this question yet.

    Since the difference between patients on corticosteroids vs not on corticosteroids is so large (HR = 0.54; near doubling of overall survival; an effect much larger than the effect of the drug itself), this becomes a very important question that needs to be answered very quickly. If corticosteroids is harmful to patients, this might be one of the reasons why the control patients on this trial did much better than the abiraterone trial as most control patients on this trial did not use corticosteroids.

    However, as you said, the entire difference might be explained as predictive correlation as patients using corticosteroids might be at a more advanced stage of the disease.

    I don’t know

  14. Dear Summer:

    These results are still (in my understanding) interim as opposed to “final” data and do not include all data on all patients prior to cross-over of men on placebo to enzalutamide. I could be wrong, but that was my understanding.

  15. Summer:

    The answer to this question will be simply resolved by the results from the PREVAIL trial of enzalutamide vs. placebo in men with chemotherapy-naive mCRPC. It is highly unlikely that any men in the control arm of that trial will be on corticosteroid therapy. By comparison, every man taking abiraterone acetate in the comparable abiraterone trial (reported at ASCO earlier this year) will have, of necessity, been taking prednisone.

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 )

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: