Ipataserib + abiraterone in the treatment of post-chemo mCRPC

One of the more provocative papers to be presented this year at the annual meeting of  the American Society of Clinical Oncology is by De Bono et al. from a large, randomized Phase II trial of a drug called GDC-0068 or ipataserib.

Ipataserib is a new type of drug that can be used to block the so-called PI3K/Akt signaling pathway, which is often highly activated in men with metastatic, castration-resistant prostate cancer (mCRPC). De Bono et al. (abstract no. 5017) combined this new drug with abiraterone acetate in the treatment of men with mCRPC who had already been previously treated with docetaxel-based chemotherapy. And they tried using two different doses of ipataserib — at 200 mg/day and 400 mg/day (see here for the summary trial protocol).

Patients were enrolled into one of three arms of this double-blind, randomized clinical trial:

  • Patients in Group A were treated with abiraterone acetate (Zytiga) at 1,000 mg/day + prednisone 10 mg/day + a placebo.
  • Patients in Group B were treated with abiraterone acetate + prednisone + ipataserib 400 mg/day.
  • Patients in Group C were treated with abiraterone acetate + prednisone + ipatasersib 200 mg/day.

The patients were also stratified to the three different trial arms based on things like the following:

  • Whether previously treated with enzalutamide (yes/no)
  • Number of chemotherapies (one/more than one)
  • Disease progression (PSA only/other).

The primary study endpoint was radiographic progression-free survival (rPFS), and a key secondary endpoint will be overall survival (OS), but the full OS data aren’t available as yet.

Here is what we know so far:

  • 253 patients were randomized to the trial.
  • 173 patients (68 percent) had had rPFS events at the time of the primary analysis.
  • Average (median) time for rPFS at the time of the primary study endpoint were
    • 6.4 months for the patients in Group A
    • 8.2 months for the patients in Group B (HR = 0.75 compared to Group A; p= 0.17)
    • About 6.5 months for the patients in Group C (HR = 0.94 compared to Group A; p= 0.75)
  • About 104 patients (41 percent) had died at the time of the primary study endpoint.
  • Average (median) OS at that time was
    • 15.6 months for the patients in Group A
    • 18.9 months for the patients in Group B (HR = 0.72 compared to Group A; p= 0.22)
    • About 15,7 months for the patients in Group C (HR = 0.96 compared to Group A; p= 0.87)
  • Adverse events more common in Groups B and C included diarrhea, nausea, vomiting, asthenia, rash, decreased appetite, and hyperglycemia.
  • The adverse events were dose-dependent, manageable, and reversible with the dose intensity of abiraterone acetate being maintained.

Thus, De Bono et al. conclude that ipataserib (400 mg/day) + abiraterone acetate + prednisone may be able to extend survival in men with mCRPC after treatment with docetaxel-based chemotherapy (although this benefit does seem to come with some risk for significant side effects).

A key question will be whether the benefit can be tied to specific and identifiable subsets of patients (e.g., those with a decreased level of PTEN expression). It is possible that more information on this will be available at the actual meeting.

The other, longer-term question will be whether using ipataserib earlier in well-defined patients may also be able to delay progression longer (perhaps withe less risk for side effects since the patients may be healthier).

6 Responses

  1. I’m confused. All the p-values are > 0.05, which means that none of the hazard ratios are statistically significant. Why would the authors then conclude there is a benefit? (I’m asking rhetorically.)

  2. Allen:

    A benefit exists if a p-value is < 0.5 not < 0.05. And an OS benefit of about 3 months would definitely be significant if it still exists after 50% of the patients have died.

  3. Sitemaster,
    A p value of 0.5 means that the observed statistic (hazard ratio, in this case) could have come about just by chance with 50% confidence. This arises because we are only looking at a limited sample of the larger population. Researchers universally use a p value of less than 0.05 (or 95% confidence) for statistical significance, meaning there is less than a one in twenty chance that the difference could have arisen purely by accident. If you want to learn more about this, see: https://en.wikipedia.org/wiki/Statistical_significance

  4. As for your other statement about the significance after half the patients have died, I think you are confusing the two quite different meanings of the word “significance.” The one I dealt with above is statistical significance, which conventionally means that we have greater than 95% confidence (p< 0.05) that the difference isn't just an artifact of random sampling. The other is a question of whether the difference is clinically meaningful. If, for example, there were only a statistically significant survival difference of a few days, we would nevertheless dismiss it because it didn't provide a meaningful benefit. Sometimes researchers define that ahead of time. So statistical significance is always a necessary pre-condition, but is not always sufficient to judge a benefit as worthwhile.

  5. Allen:

    Nowhere did I or the authors suggest that this study meets levels of statistical significance as you are (correctly) defining them. This trial wasn’t designed to do that, and no one is going to be approving ipataserib on the basis of this trial. But a 3-month survival benefit in a trial like this is still “significant” and (unless the side effects of treatment are actually more severe than suggested by the report) is likely to men that this drugs will move into Phase III trials.

    I was not “confused” at all. You were imputing something that Neither I nor the authors had actually said. :O)


  6. Mike:

    Statistical significance is the most basic requirement for ALL medical research, else why bother recruiting a large number of patients for a trial. It doesn’t have to be said. This was a Phase II trial designed to test safety and efficacy. It’s primary efficacy goal was radiographic progression-free survival, and secondary efficacy goals included overall survival, PSA response, RECIST score changes, time to progression, CTC, and pain scores. It did not meet its goals. All medical research begins with the “null hypothesis” that the treatment makes no difference. That hypothesis is rejected (i.e., the treatment had an effect) if and only if the effect is observed with at least 95% confidence (p< 0.05).

    The sponsor may decide to take a gamble and move the drug to Phase III trials in spite of failing to demonstrate any statistically significant benefit in Phase II trials, but that is a big risk that their board of directors may look askance at. When the authors state, “This study provides evidence that Ipat in combination with Abi may improve rPFS and OS in pts with mCRPC post docetaxel” they are very much mistaken — this study provides no such evidence because it failed to meet the basic requirement of statistical significance. I’d be very much surprised if this gets published in a peer-reviewed journal as it is written.

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: