Wherefore Art Thou, Provenge? - Charles Ryan

A meditation on the need and utility for Sip T in 2018. On a nearly weekly basis I am confronted by the question of whether I should be giving a patient Sipuleucel T. I still use it, but not in all my CRPC patients, and I like to think of this therapy as having a reasonable benefit for patients - who are at an early point in the CRPC spectrum. I will explain but first let’s review the data and its limitations.

Sipuleucel T was approved by the FDA because of its effect on the overall survival of a population of CRPC patients. The study was published in 2010 in the New England Journal of Medicine (1). The trial, called IMPACT, was a randomized placebo controlled trial with a 2:1 allocation to Sip T or the infusion of a placebo stimulated cellular product. Leukapheresis-acquired cellular samples were stored in the placebo treated patients, and made available to them for crossover treatment at disease progression in those who requested it.

Median survival was 25.8 months in the Sip T arm and 21.7 in the placebo arm. More importantly however, was the fact that the hazard ratio for death was 0.775, indicating that receiving Sip T reduced the risk of death by about 23% over time. That is not a huge change but it is statistically significant. Is it clinically significant as well?

Here are a few key points about the IMPACT study that are worth considering in 2018:

  1. The study was performed before we had Abiraterone, Enzalutamide, Apalutamide, Radium 223 or Cabazitaxel and thus may not accurately reflect the natural history of the disease in 2018. The data lock for IMPACT was in 2009, the above-mentioned therapies were all approved after 2010.
  2. The majority of patients (80-85%) were chemotherapy naïve.
  3. Crossover was allowed. Forty nine percent of the patients initially randomized to the placebo arm received Sip T as their first subsequent treatment following disease progression and 64% received Sip T “at some point” subsequently. This fact makes the data more impressive, suggesting that early Sip T is superior to later administration.

A subsequent publication (2) evaluated the effect of Sip T on time to disease related pain and time until first opioid administration. The former analysis, time to disease related pain, is considered not significant, however one look at the Kaplan Meier plot demonstrates that, although they are not different when evaluating the median, it is likely that Sip T delayed pain and symptoms in a subset, supporting the notion that Sip T does a lot of good for some, but may not work in others. And, to quote Hamlet, there’s the rub.

After administering Sip T to many patients now over the last 8 years or so, I’ve come to see that that data point is likely correct. It is hard to tease out sip T’s effects on an individual patient and so, frequently, I will simply give the treatment, declare victory, and move on. Frustratingly, although a number of studies have looked at T cell activation as a biomarker of activity, such tests are not validated as a means of determining who is ‘victorious’ and is benefitting from the therapy.

But one of the values of Sip T as I see it is not in its survival benefit but rather the benefit that may be present in delaying the critical need for a subsequent therapy, thus saving money and exposure the side effects of treatment. I think that the survival effect may get lost in patients treated in 2018, given the efficacy of all of the subsequent treatment. It’s also entirely possible that the time to opiate analgesic benefits would get negated by the use of other treatment. I don’t watch patients get worse until they need an opiate. I start other treatments hopefully long before that happens based on PSA rise, scans and other factors. Thus by the time a patient gets to the point of needing an opiate, more than one treatment has failed.

Given the availability of Sip T and making decisions about when and if to use it always allow me to think about the spectrum of CRPC and where the therapy fits. While considering the use of the therapy I ask myself a few quick questions that, in my view, frame the appropriate moments in which Sip T can be given.

  1. Is this patient asymptomatic and with a low volume of disease?
  2. Is this patients prognosis approximately 3 years or longer?
  3. Would it be acceptable to me and the patient to NOT get any other treatment (e.g. abiraterone or enzalutamide) for a period of time of 6-12 months?

Put another way, my main reason to use Sip T these days is to ensure the indolence of already indolent disease. That’s right, my perspective on the data from the Impact study is that it improved the prognosis of those who already had a good prognosis. It didn’t reverse aggressive disease. The magnitude of benefit in the IMPACT study was greatest in those who lived three years. The three-year survival was increased by about 35-40% compared to placebo, even though median survival was only increased by 23%. The three-year survival for Sip T treated patients was 32% versus 23% for those who initially received placebo.

So what do I do after I’ve given Sip T to a patient? I follow him closely. If I don’t think I can follow the patient and I don’t think that it is going to be safe to do so (e.g because their disease is progressing so quickly) I simply won’t give the Sip T- I’ll move on to abiraterone, enzalutamide or chemotherapy.

So, where do we go from here? Giving Sip T now comes with a mild feeling of dissatisfaction. I wish I knew who was going to benefit, and I wish we were building on this treatment by enhancing its efficacy with either combination approaches, treatments that may augment antigen presentation or may enhance effector T cell action against prostate cancer. Studies are underway on a small scale, but to my knowledge there are no phase III studies underway or planned in which Sip T is integral. I look forward to seeing such studies develop.

Written by: Charles J. Ryan, MD

1. Kantoff Et al N Engl J Med. 2010 Jul 29;363(5):411-22.
2. Small et al Prostate Cancer and Prostatic Disease 2014 Sep;17(3):259-64

Read More:
Survival Outcomes For African-American Versus Matched Caucasian Patients With mCRPC Treated With Sipuleucel-T