Keeping progress in perspective — a message to Prof. Alan Ashworth


With the publication of data about enzalutamide in the New England Journal of Medicine last week, Professor Alan Ashworth, the head of the Institute for Cancer Research (ICR) in the UK, proclaimed “a golden age in prostate cancer drug discovery.”

Prof. Ashworth’s enthusiasm is certainly understandable. No one is going to disagree with him that great progress has been made in the potential to treat progressive forms of prostate cancer in the past few years:

  • In April 2010, sipuleucel-T became the first form of personalized immunotherapy ever to extend overall survival in cancer (for men with asymptomatic or minimally symptomatic CRPC — metastatic or non-metastatic).
  • Cabazitaxel — a taxane-based chemotherapeutic agent — used in combination with prednisone was approved (in June 2010) to extend overall survival in men with metastatic, castration-resistant prostate cancer (mCRPC) progressing after docetaxel-based chemotherapy.
  • Abiraterone acetate + prednisone initially showed an overall survival benefit in men with mCRPC after docetaxel-based chemotherapy (and was approved for that use in April 2011) and — more recently — has shown a progression-free survival benefit in chemotherapy-naive men with asymptomatic or minimally symptomatic mCRPC.
  • Radium-223 is the first injectable form of radiation therapy to demonstrate an overall survival benefit in any form of cancer (specifically in men with mCRPC who had or had not received prior chemotherapy).
  • Enzalutamide (formerly known as MDV3100) has demonstrated an overall survival benefit in men with mCRPC after docetaxel-based chemotherapy and is still expected to demonstrate an overall survival benefit in chemotherapy-naive men with asymptomatic or minimally symptomatic mCRPC.

However, among these five new agents, the longest median overall survival benefit is still “only” 4.8 months (demonstrated by men in the enzalutamide trial). In that trial, the median age of the 800 men randomized to receive active drug was 69 years (range 41 to 92 years). Thus, the average extension of life of these patients was actually

4.8/(69 x 12) x 100 = 0.58 percent

The way that The “New” Prostate Cancer InfoLink looks at the progress made over the past 5 years or so is not so much as “a golden age” in prostate cancer drug discovery as it is a platform for real hope that such a golden age may be coming.

For us, that “golden age” will only begin when new forms of therapy start to show a median survival benefit that extends life by at least a year in men with mCRPC. In the enzalutamide trial mentioned above, a median overall survival benefit of a single year would have equated to a 1.45 percent extension in the overall life of the men on active drug. However, very few forms of cancer therapy have ever demonstrated a median overall survival benefit of a year for people with advanced, metastatic disease, so this would be a really meaningful change in the opportunity available to patients — especially if that benefit was shown by a form of therapy that had the potential to be used much earlier in the progression of the disease (and therefore to induce a far greater survival in less sick and much younger patients).

At present, we know of just one agent in advanced clinical trials that might even begin to approach this type of overall survival benefit — the immunotherapeutic agent Prostvac VF, which, in a randomized Phase II clinical trial, showed an overall survival benefit of 8.2 months. However, overall survival was not the primary endpoint of the trial in question, and so this result needs to be treated with great caution.

Of course the true “golden age” for research into treatment of prostate cancer will only begin if we can start to find ways to absolutely stop progressive prostate cancer in its tracks, without most of the side effects historically associated with things like surgery and radiation therapy. However, when  we consider the progress made over the past 5 to 10 years, even that golden age may be closer than we could possibly have believed just 20 years ago.

6 Responses

  1. Fact is overall survival data like 4.8 months is a median. In the Provenge trial some patients lived for more than 3 years. Also in the Provenge trial, patients in the placebo arm received (were crossed over to) frozen Provenge from the
    beginning. It was a disadvantage for the company. Dendreon had to compete against its own product in a frozen condition. And still Provenge reached a median overall survival benefit of 4.1 months. The actual median survival benefit may be as much as 7.8 to 12.1 months.

    When you break the data from the 512-patient study into subsets based on baseline PSA, the results, in the words of Chodak et al., “… support a consistent overall survival, or OS, benefit with sipuleucel-T across PSA quartiles.” As noted by the authors, in the IMPACT study (n = 512), there was a 4.1-month improvement in median OS (25.8 vs. 21.7 months) for sipuleucel-T compared with control (HR = 0.78; 95% CI: 0.61, 0.98; p = 0.032). Importantly, 109/171 (64%) of control patients received APC8015F(frozen sipuleucel-T). “Post-progression treatment with APC8015F may have extended the survival of control subjects in the IMPACT study. Adjusting for use of APC8015F resulted in an increase in median OS benefit with sipuleucel-T of up to 7.8 months. These results suggest a greater treatment effect of sipuleucel-T than reported in IMPACT, and should be factored into future studies without APC8015F crossover.”

    Click here for the abstract of the paper presented by Gomella et al. at the AUA annual meeting this year and click here for the full poster.

  2. I probably shouldn’t quibble, since I’m always wishing for as much specificity and quantification as possible, but …

    This is one of those cases where the median is less helpful than the mean would be. Although common sense tells us that your result (0.58 percent) is probably close to the truth for any sensible data set, and that 0.6 percent would be closer, we cannot calculate any such exact value from the data provided; they are equally consistent with a fantasy where half the men died on the spot and the rest became immortal.

    Nonetheless, I agree with your overall point: A brief postponement of an inexorably advancing death-by-prostate-cancer is an admirable goal, but the interests of public health and quality-of-life are better served by treatment in earlier stages, when it might provide a decade-long postponement or even a cure.

  3. Dear John:

    First, you are completely correct, the median survival benefit of men initially treated with sipuleucel-T in the IMPACT trial was 4.1 months compared to the men initially treated with a placebo. No one has ever suggested anything else. The median overall survival of the men initially treated with sipuleucel-T in the IMPACT trial was 25.8 months compared to a median overall survival of 21.7 months for the men initially treated with a placebo.

    Second, the hypothesis that the crossover of patients who had originally received a placebo to receive the frozen product properly known as APC8015F in the IMPACT trial is just that — an hypothesis, and I have no reason to believe that Dr. Gomella and his colleagues believe otherwise. There are no other data (that I am aware of) to support this hypothesis. Hypotheses like this based on retrospective data analysis are scientifically interesting, but they are not medical proof of anything.

    I would also note that:

    (a) The crossover of patients to APC8015F was voluntary and occurred at the time of their progression (in some patients) and later on (in other patients, after they had received other types of therapy).

    (b) My understanding is that it was Dendreon that proposed this protocol. I am not aware of any reason why the FDA would have ever considered the need for, let alone required, such a protocol. This may, in some way, have been associated with the fact that the IMPACT trial was an expansion of another study that Dendreon had initiated before the IMPACT trial was fully put in place.

    (c) The length of time that men lived on the trial is certainly important. If you look at the Kaplan-Meier curve in the published report of the results of the IMPACT trial (Figure 2 in the paper by Kaplan et al. in the New England Journal of Medicine) the authors suggest that: after 3 years, 49 men initially treated with sipuleucel-T were still alive compared to 19 men initially treated with placebo; after 4 years, 14 men initially treated with sipuleucel-T were still alive compared to 1 man initially treated with placebo; and after 5 years, 1 man was still alive from each group. Findings like this are not unusual in trials of this type. (Please remember that twice as many men were randomized to sipuleucel-T as were randomized to placebo.)

    Having stated all of this, I am still trying to understand the point you are trying to make. It is certainly true that if you isolate the men in the trial at lowest risk based on their PSA level, and then carry out another mathematical analysis (as carried out by Chodak et al.) that the median overall survival times and the median overall survival benefit are both longer. I think it is probably a pretty safe bet that if you isolated men at lowest risk for progression in any other similar trial, they would also show a greater median overall survival than the median survival benefit for all the men in the trial. What is the surprise? What this tells us is that men with asymptomatic or minimally symptomatic disease and with lower PSA levels are potentially better candidates for Provenge therapy. It doesn’t tell us that all men with asymptomatic and minimally symptomatic disease had their life extended by at least a year after treatment with sipuleucel-T.

  4. Paul:

    Since my only meaningful point was the overall one (and the data are only illustrative), it would seem we are in complete agreement!

    (Incidentally … The the upper limit of the survival of the men treated with enzalutamide was not known with 95% confidence at the time the randomization was broken and all patients were offered active drug. For all we know, the real overall survival benefit of enzalumtamide might have been much more than 4.8 months, but it would have been unethical to continue the trial based on the data available.)

    :O)

  5. … But it tell us that all men that receive sipuleucel-T (Provenge) have a chance for overall survival and that chance can give them more than 3 years. Ask any sick man and his family if they want this chance and certainly most of them will say “Yes”.

  6. Dear John:

    Respectfully, that is not true, what it tells us is that some men that receive sipuleucel-T have a better chance of longer-term overall survival than others, and that the probability of that longer-term overall survival appears to depend on a number of factors (such as their PSA and the degree of spread of the metastasis at the time they start taking this agent). No one actually knows whether the use of APC8015F had any effect at all on the survival data of patients in this trial, so the idea that treatment with sipuleucel-T will be likely to extend the survival of most patients by something like 6 months to a year (compared to doing nothing) is factually inaccurate.

    Hope is a wonderful thing. Unreasonable hope based on a potentially mistaken belief is a delusion.

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: