Is treatment with degarelix really “better” than treatment with an LHRH agonist? Part II


Yesterday we commented (by request) on a paper by Klotz et al. suggesting that initial treatment with the LHRH antagonist degarelix might have benefits compared to initial treatment with an LHRH agonist such as leuprolide or goserelin acetate.

We have now had the chance to read (several times) the full text of the original paper by Klotz et al. Having done so, all we can really tell readers is that while the data presented in this paper are almost certainly statistically accurate, the clinical significance of these data appears to be of debatable if not dubious validity. In what follows we have attempted to explain the details of our concerns (above and beyond those we had previously set out).

First, let us be very clear that the meta-analysis conducted by Klotz et al. was, in fact, based on pooled, individual patient data from the five trials specified in the paper. This is a strength of the analysis.

Second, let us also be very clear that neither prostate cancer progression (biochemical or clinical), nor prostate cancer-specific survival, nor overall survival were the primary endpoint in any of the five studies included in the meta-analysis. Furthermore, it appears that no patient in any of the five studies was actually followed for more than 1 year, which means that there are major limitations to the clinical validity of conclusions about progression-free and other survival data suggested in the meta-analysis.

Third, there are some very specific methodological findings that are evident in the full text of the paper:

  • The five trials (of which the largest apparently remains unpublished) can basically be divided into two groups.
    • In two of the trials (see Klotz et al. and an unpublished study by Shore et al.), which included 1,458 patients in total, the patients were followed on therapy for 12 months in total.
    • In the other three trials (see Anderson et al., Mason et al., and Axcrona et al.), which included 467 patients in total, the patients were followed on therapy for just 3 months.
  • “PSA PFS was defined as death or PSA recurrence”. In other words, a man who died within a week of starting treatment with degarelix or an LHRH agonist was considered to be equivalent to a man who might have a rising PSA (either two consecutive increases in his PSA level of ≥ 50 percent compared to his nadir PSA level or a rise of ≥ 5 ng/ml on two consecutive PSA assays from blood drawn at least 2 weeks apart) at any time up to 364 days after initiation of therapy.

The “New” Prostate Cancer InfoLink claims no great expertise in the validity of the conduct of meta-analyses, but it would seem to us that:

  • The consolidation of data from patients who were followed for 12 months together with patients followed for just 3 months in a meta-analysis of this type was questionable.
  • The definition of biochemical progression-free survival described above is at best “unusual”.

If we now move to look at how the results of the meta-analysis are presented, there are further oddities.

  • Figure 1 shows two Kaplan-Meier plots (the classic method by which to chart cancer survival data) in which fewer than 25 percent of patients exhibit disease progression during the 364-day follow-up period.
  • Figure 2 shows another Kaplan-Meier plot in which fewer than 5 percent of patients died within 364 days of follow-up.

Under the normal “rules of the road” for the use of Kaplan-Meier techniques to suggest a progression-free or overall survival benefit, one expects at least one curve on a chart to show that at least 50 percent of the patients in a specific subset had progressed or died within the follow-up period.

If we look at the side effect data reported in the full text of the paper, the following also become clear:

  • Adverse events of any type were reported in
    • 942/1,266 patients treated with degarelix (74 percent)
    • 445/659 patients treated with an LHRH agonist (68 percent)
  • Hot flashes are the most commonly recognized and aggravating side effects of androgen deprivation therapy reported by patients; in the data presented in this paper, hot flashes were reported by
    • 386/1,266 patients treated with degarleix (30 percent)
    • 171/659 patients treated with an LHRH agonist (26 percent)
  • The most common class of adverse events, by far, were injection site reactions at or after treatment, which occurred in
    • 786/1,266 patients treated with degarelix (62 percent)
    • 6/659 patients treated with an LHRH agonist (1 percent)

In the abstract of the paper, however, the authors single out “fewer joint-related signs and symptoms, musculoskeletal, and urinary tract events in the degarelix group.” We would note that these types of adverse reactions occurred in less than 5 percent of patients reporting side effects associated with treatment with either degarelix or an LHRH agonist, i.e., fewer than 5 percent of the 1,387 patients from whom side effect data were available, which is at most 69 patients. In saying this, we do not wish to imply that such side effects are unimportant. Clearly they are. However, among all the common adverse reactions reported as occurring in > 5 percent of patients, not one was more common in patients being treated with degarelix compared to an LHRH agonist.

Finally, with respect to actual deaths occurring in the five trials:

  • 18/1,263 patients (1.4 percent) died while on treatment with degarelix and three of those patients (0.2 percent) actually died of prostate cancer.
  • 19/657 patients (2.9 percent) died while on treatment with an LHRH agonist and just one of those patients (< 0.2 percent) actually died of prostate cancer.

We are going to leave it to others to decide for themselves what this paper “means” in  terms of the relative benefits of degarelix compared to an LHRH agonist in the initial treatment of progressive prostate cancer. There are two things only that we feel quite sure about:

  • Degarelix is, quite certainly, capable of lowering the PSA levels and the serum testosterone levels of men with prostate cancer faster and (sometimes) further into the undetectable range than the LHRH agonists, and this may be important and beneficial for some men.
  • Degarelix is associated with a very significantly higher risk for injection site reactions to treatment than either goserelin acetate or leuprolide acetate

Editorial comment: Our thanks to a primary care physician who is also a prostate cancer patient in The Netherlands for providing us with the full text of this paper by Klotz et al.

One Response

  1. Now that you have exposed the details, it seems irresponsible and misleading for the those in charge of the study to have given the impression that degaralix showed a 53% overall survival benefit compared to an LHRH agonist, without noting in bold print the limitations of the results. This shows again that it is “buyer (patient) beware” when relying on a study’s conclusion without criticially reviewing the underlying data. Thanks again Sitemaster for shining the light!

    Richard

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: