Sipuleucel-T (Provenge®) for castration-resistant prostate cancer


  • Giuseppe Di Lorenzo,

    Corresponding author
    1. Cattedra di Oncologia Medica, Dipartimento di Endocrinologia e Oncologia Molecolare e Clinica, Università degli Studi Federico II, Napoli, Italy
      Giuseppe Di Lorenzo, Cattedra di Oncologia Medica, Dipartimento di Endocrinologia e Oncologia Molecolare e Clinica, Università degli Studi Federico II, Naples, 80131, Italy. e-mail:
    Search for more papers by this author
  • Matteo Ferro,

    1. Cattedra di Oncologia Medica, Dipartimento di Endocrinologia e Oncologia Molecolare e Clinica, Università degli Studi Federico II, Napoli, Italy
    Search for more papers by this author
  • Carlo Buonerba

    1. Cattedra di Oncologia Medica, Dipartimento di Endocrinologia e Oncologia Molecolare e Clinica, Università degli Studi Federico II, Napoli, Italy
    Search for more papers by this author

Giuseppe Di Lorenzo, Cattedra di Oncologia Medica, Dipartimento di Endocrinologia e Oncologia Molecolare e Clinica, Università degli Studi Federico II, Naples, 80131, Italy. e-mail:


What's known on the subject? and What does the study add?

Sipuleucel-T is known to be very well tolerated and to prolong overall survival, but not progression-free survival, measured according to prostate-specific antigen variations and radiographic progression. Its exact mechanism is unknown.

Although the article does not present new data, a new way of assessing immunotherapy efficacy is proposed. This involves measuring ‘consecutive’ times to progression, in an attempt to capture the delayed effects of immunotherapy.


  • • To propose a new way of assessing immunotherapy efficacy. Since 2010, the therapeutic armamentarium for prostate cancer has expanded to include the potent taxane agent cabazitaxel, the CYP17A1 inhibitor abiraterone and the novel immunotherapy agent sipuleucel-T (Provenge®. Demdreon, Seattle, WA, USA).
  • • Sipuleucel-T is an antigen-specific active immunotherapy agent, which is not designed to be directly toxic to tumour cells, but to help the immune system to selectively attack cancerous cells. We aimed to provide a comprehensive review of available safety and efficacy data about Sipuleucel-T.


  • • A systematic analysis of the literature was conducted using the terms ‘Sipuleucel-T’ and ‘Provenge’. PUBMED was the main search engine, but abstracts published by the American Society of Clinical Oncology and the European Society of Medical Oncology, as well as press releases and product monographs, were also considered for inclusion.
  • • Reference lists of key articles were searched for further leads. Articles providing safety and efficacy data were included in this review.


  • • Sipuleucel-T is based on autologous dendritic cells, which are collected by leukapheresis of peripheral blood, co-cultured with a modified PAP protein, and then re-injected intravenously. It is the first agent of its kind to obtain Food and Drug Administration approval for any kind of malignant tumour.
  • • Its approval was determined by the results of a placebocontrolled, randomized trial (the IMPACT trial), conducted in 512 asymptomatic or minimally symptomatic mean with metastatic castration-resistant prostate cancer.
  • • Although no difference in time to progression or PSA response rate was reported, a statistically meaningful 4.1-month improvement in median survival was achieved in the active arm with respect to the placebo arm (25.8 months vs 21.7 months).
  • • After Food and Drug Administration approval in April 2010, in view of the high economic cost of sipuleucel-T and the not completely flawless study design of the IMPACT trial, a national coverage analysis of sipuleucel-T was conducted by the Centers for Medicare and Medicaid Services. Such analysis has recently concluded that sipuleucel-T is a ‘necessary and reasonable’ treatment.


  • • Sipuleucel-T is an effective treatment for prostate cancer, although its widespread use is uncertain for complex social and economic reasons.

castration-resistant prostate cancer


prostatic acid phosphatase


granulocyte–macrophage colony-stimulating factor.


Great efforts to provide a benefit in survival as well as in quality of life for patients affected by castration-resistant prostate cancer (CRPC) have allowed three effective options to be added to the therapeutic armamentarium for CRPC, which had been limited to docetaxel and prednisone alone in the past two years. In the past years, the mature results of three positive phase III trials of three novel agents belonging to different pharmacological classes have become available [1]. Two of these agents – abiraterone, an orally administered CYP17A1 inhibitor, and cabazitaxel, a potent, intravenously administered taxane – are pharmacodynamically similar to other drugs currently employed for CRPC, that is to say ketoconazole and docetaxel, respectively. The third agent, sipuleucel-T (Provenge®, Dendreon, Seattle, WA, USA), is the first vaccine to be approved for CRPC with no, or minimal, symptoms [1]. Among solid malignancies, prostate cancer appears particularly suitable for an immunotherapy-based approach because of several peculiar features that distinguish it from other solid malignant tumours. These features include its often slow cancer progression, which can span more than a decade, from localized castration-sensitive to metastatic castration-resistant disease. It is also characterized by an ineffective spontaneous immune response, which can be enhanced by hormonal manipulations, as well as the existence of tumour-specific and tissue-specific antigens, such as prostatic acid phosphate (PAP) and PSA, which can be targeted without the risk of autoimmunity [2]. Sipuleucel-T is classified as active immunotherapy, in that, unlike monoclonal antibodies, which directly target specific antigens or molecules, sipuleucel-T sensitizes the patient's immune system against specific antigens or molecules, to train the system to attack the tumour [3]. Sipuleucel-T accomplishes this goal by using patient's autologous specialized antigen-presenting cells, i.e. dendritic cells [3]. In contrast to other forms of antigen-specific, active immunotherapy, such as whole cell-based, DNA-based and peptide-based agents and viral vectors, autologous dendritic cells provide a potent, antigen-specific stimulation of both T and B cells, with no risk of an inappropriate immune reaction against the vaccine components, given the autologous nature of the cells employed [4].

The IMPACT trial [5] is the largest of three randomized, placebo-controlled, double-blind trials [5,6] that have concordantly indicated a survival advantage in asymptomatic or minimally symptomatic patients with metastatic CRPC treated with sipuleucel-T, in the absence of any difference in PSA or radiographic response rate or progression-free survival. In view of the high cost of sipuleucel-T , coupled with a not flawless study design [7], the approval of sipuleucel-T by the Food and Drug Administration in April 2010 was followed by the opening of a national coverage analysis for sipuleucel-T by the Centers for Medicare and Medicaid Services in June 2010, to determine whether the drug was ‘necessary and reasonable’[8].

This review discusses the clinical evidence available for sipuleucel-T, with a focus on methodological and socio-economic issues judged of particular interest.


To manufacture sipuleucel-T, a patient's own dendritic cells are collected by leukapheresis of peripheral blood, and then co-cultured with PA2024, a fusion protein of PAP, and granulocyte–macrophage colony-stimulating factor (GM-CSF) [3] (Fig. 1). The PAP component of the PA2024 protein serves to specifically direct the immune response against prostate cancer, considering that PAP is universally expressed by and largely limited to prostatic cells, whereas the GM-CSF component serves to enhance the immune response. During ex vivo culture, autologous dendritic cells take up and process the PA2024 protein into small peptides, which are then combined with MHC II molecules and displayed on the cell surface. Apart from presenting the processed PA2024 antigen, dendritic cells must express additional co-stimulatory molecules on the cell surface for an effective stimulation of CD4+ cells – a process which is called ‘activation’. Each dose of sipuleucel-T contains at least 50 million autologous dendritic cells activated with PA2024, as ascertained by CD54 expression, which is a reliable marker of immune cell activation. The final product also contains a variable number of T cells, B cells, natural killer cells and other cells [3]. As specified in the Food and Drug Administration-approved package insert, sipuleucel-T must be administered intravenously over a period of approximately 1 h, with a post-infusion observation period of at least 30 min. As sipuleucel-T can cause infusion reactions, patients should be orally premedicated with acetaminophen and an antihistamine drug such as diphenhydramine approximately 30 min before the infusion. According to the recommended schedule, three complete doses of sipuleucel-T are administered at approximately 2-week intervals. In controlled clinical trials, the dosing interval ranged from 1 to 15 weeks, but the maximum allowed dosing interval has not been established [9].

Figure 1.

Manufacture of sipuleucel-T (Provenge®); reproduced by kind permission from Dendreon.


The most comprehensive analysis [9] carried out on sipuleucel-T safety included 904 patients, receiving sipuleucel-T (601 patients) or placebo (303 patients), who were enrolled in four randomized controlled trials, that is to say the completed IMPACT [5], D9901 [6] and D9902A [6] trials and the currently ongoing PROTECT trial (see Table 1). Sipuleucel-T was well tolerated, with chills, fever, muscle ache, asthenia, headache, vomiting and pain of any grade reported in more than 10% of patients in the active arm, and more frequently with sipuleucel-T than with the placebo. Grade 3 and grade 4 adverse effects respectively occurred in 23.6% and 4.0% of patients in the sipuleucel-T group compared with 25.1% and 3.3% of patients in the control group, while grade 3–5 chills, headache and hypertension occurred in 0.5–2.2% of patients given sipuleucel-T and in no patient in the control group. Cerebrovascular events were more frequent in patients receiving sipuleucel-T (3.5% vs 2.6%), but the absolute risk difference was too small to draw any conclusion [9]. Another safety issue of interest concerns the risk of infection, carried by the possible bacterial contamination of sipuleucel-T during its manufacture. Sipuleucel-T is routinely tested for microbial contamination, but the fact that sterility tests are not fully available until after the infusion warrants careful monitoring of patients for any sign of infection. The 5% incidence rate of infections deemed to be infusion-related was similar in the active and placebo arms, although it must be taken into account that in the latter group patients were not given an inert placebo, but received intravenous injection of non-activated autologous peripheral blood mononuclear cells, which are also subject to the risk of procedure-related contamination [9].

Table 1.  Ongoing trials on sipuleucel-T for prostate cancer identifierTrialMain inclusion criteriaStudy design and protocolEnd points
  1. CRPC, castration-resistant prostate cancer; FDA, Food and Drug Administration; PSA, prostate-specific antigen.

NCT00715104Sipuleucel-T as Neoadjuvant Treatment in Prostate CancerMen with adenocarcinoma of the prostate scheduled for radical prostatectomy as the initial therapy for localized prostate cancerPhase II, single arm, 40 patients to be accrued.Immune response within prostate tissue after neoadjuvant treatment with sipuleucel-T and toxicity
Patients will receive neodjuvant sipuleucel-T before prostatectomy
NCT00779402Provenge for the Treatment of Hormone Sensitive Prostate Cancer (PROTECT)Men with adenocarcinoma of the prostate who have undergone radical prostatectomy as the initial therapy for localized prostate cancer, presenting with castration-sensitive PSA-raising only diseasePhase III, two arms, double blind, 159 patients to be accrued.Time to biochemical failure, time to distant failure, PSA doubling time, overall survival
Patients will receive either sipuleucel-T or placebo in combination with a short-term androgen deprivation therapy
NCT01306890A Registry of Sipuleucel-T Therapy in Men With Advanced Prostate Cancer (PROCEED)Men with CRPC receiving Sipuleucel-T according to its FDA-approved useProspective observational study, 1500 patients to be enrolled. Patients will receive sipuleucel-TIncidence of cerebrovascular events, overall survival
NCT01274572Blood for Immune Response to Provenge® in HRPCMen with CRPC receiving sipuleucel-T according to its FDA-approved useProspective observational study, 50 patients to be enrolled. Patients will receive sipuleucel-T according to its FDA-approved use, with the collection of additional blood samplesImmune response
To Evaluate Sipuleucel-T Manufactured With Different Concentrations of PA2024 Antigen (ProACT)NCT00715078Men with asymptomatic or minimally symptomatic metastatic CRPCPhase III, three arms, single blind, 150 patients to be accrued.CD54+ upregulation, levels of immune response, overall survival, circulating tumour cells
Patients will receive sipuleucel-T manufactured with one of three different concentrations of the PA2024 antigen

The three phase III trials [5,6] conducted in asymptomatic or minimally symptomatic patients with metastatic CRPC were similar in study design, inclusion and exclusion criteria, and results. Docetaxel use was not forbidden nor required, and it could be administered on disease progression at the investigator's discretion. The main exclusion criteria were an Eastern Cooperative Oncology Group performance status of 2 or more, the presence of visceral metastases, pathological long-bone fractures or spinal cord compression. Patients who had received treatment with systemic glucocorticoids within the previous 28 days or with chemotherapy within the previous 3 months were not eligible for the trials. Patients with a Gleason score higher than 7 were initially excluded from the IMPACT trial, but could be enrolled after a protocol amendment [5].

In all of the three trials, antigen-presenting cells were collected from patients in both groups and then manufactured and administered to the active arm as described earlier in this review. One-third of antigen-presenting cells collected from patients in the placebo group were injected without being pulsed with PA2024, whereas the remaining two-thirds were cryopreserved to generate the open-label product that could be administered at progression to patients of the placebo arm. Median time to disease progression was the primary endpoint of the D9901 and of the D9902A [6], whereas overall survival was the primary endpoint of the IMPACT trial [5]. In the IMPACT trial, which randomized a total of 512 patients to sipuleucel-T (n = 341) or placebo (n = 171), a relative reduction of 22% in the risk of death was reported in the group given sipuleucel-T, with respect to the placebo group (hazard ratio 0.78; 95% CI 0.61–0.98; P = 0.03, log-rank, Fig. 2). In the combined analysis of the D9901 and D9902A [6] trials, a total of 225 patients were randomized to sipuleucel-T (n = 147) or placebo (n = 78), with a 33% reduction in the risk of death (hazard ratio 1.50; 95% CI 1.10–2.05; P = 0.011, log-rank). As anticipated, no signal for improved response-based or progression-based endpoints was reported in any of these three trials. One major bias may originate from subsequent docetaxel, not defined by protocol. In the IMPACT trial, a slightly greater proportion of patients enrolled in the active arm received post-treatment docetaxel (about 57% vs 50%). Although sensitivity analysis revealed no influence of docetaxel in the IMPACT population on survival, such a conclusion inevitably has important limitations, in view of its retrospective nature, the lack of data regarding dosing and scheduling of docetaxel, as well as any other treatment of potential efficacy that patients could receive in the post-progression, open-label phase of the trial. Nevertheless, an integrated analysis of the IMPACT, the D9901 and the D9902A trials [10], recently published in abstract form, confirmed the absence of docetaxel influence on survival. In fact, by pooling the data regarding 737 patients, sipuleucel-T treatment was associated with a 26.5% reduction in risk of death (hazard ratio 0.735, 95% CI 0.613–0.882, P < 0.001), and the sipuleucel-T effect was also significant if patients were censored at the time of docetaxel use (hazard ratio 0.714, 95% CI 0.562–0.908, P = 0.006, log-rank). In this combined analysis, median time to docetaxel use was 8.4 months, a timeframe that seems appropriate to avoid possible detrimental effects on the immune system by docetaxel but exploit docetaxel effectiveness.

Figure 2.

Overall survival in the IMPACT trial (study has been previously published; Kantoff et al. [5] permission granted).

The lack of a demonstrated advantage in progression or response-based endpoints has highlighted the importance of identifying non-conventional variables that can serve as surrogates for efficacy and that can be measured in the short to mid-term. Interestingly, in the IMPACT trial, titres of antibodies against PA2024 that exceeded 1:400 [5] and transient elevation of eosinophil counts [11] at any time after baseline were significantly higher in the active arm and were associated with improved survival in the overall population, whereas increased total γ-globulin levels were in turn associated with improved titres of anti-PA2024 antibodies [12]. Research has also focused on identifying measurable variables potentially indicative of product quality. Interestingly, greater levels of antigen-presenting cell activation-associated cytokines (such as interferon-γ and tumour necrosis factor-α) and T-cell activation-associated cytokines (such as interleukin-2 and interleukin-4) have been found in supernatants from antigen-presenting cell culture with PA2024, with respect to supernatants from antigen-presenting cell culture with GM-CSF alone [13]. Furthermore, an analysis involving 476 patients treated with sipuleucel-T and enrolled in the three phase III trials mentioned, found a positive and significant correlation between product parameters, such as the total number of nucleated cells and the cumulative CD54+ cell count, with overall survival [14]. These preliminary data may provide valuable input for addressing the important need to identify efficacious intermediate endpoints and valid measures of product quality.

Sipuleucel-T has also been investigated for prostate cancer patients not affected by metastatic CRPC, but no phase III trial has been completed in this setting. In a single-arm phase II trial, a signal for improved doubling time of PSA level was reported in 13 of 18 patients with castration-sensitive prostate cancer [15]. In a double-blind, placebo-controlled trial, enrolling 186 patients with PSA-raising only castration-sensitive prostate cancer, a statistically significant increase in doubling time of PSA level was associated with sipuleucel-T treatment, although no significant difference in time to biochemical failure or distant metastasis was reported [16]. In the same setting, similar results in terms of improved doubling time of PSA level were obtained with the combination of bevacizumab and sipuleucel-T [17], although failure of bevacizumab to provide a survival advantage in combination with docetaxel with respect to docetaxel alone [18] has discouraged further investigations in this direction.


The magnitude of the results for an active, antigen-specific immunotherapy agent for prostate cancer has profound clinical and scientific consequences, along with raising important social and economic issues. The Prostate Cancer Working Group 2 criteria, which moved the focus from response-based and PSA-based endpoints to progression-based and measurable disease-based endpoints [18], provided no valid tool for the evaluation of immunotherapy efficacy, which presently relies on overall survival, as in the case of sipuleucel-T. Although the ease and objectivity of its measurement make overall survival an excellent endpoint, in a scenario where immunotherapy is administered early during the course of CRPC, biases caused by inevitable imbalances in subsequent treatments are difficult to avoid [19], especially in view of the growing number of therapeutic options available for the disease [18]. The intrinsic limitation of overall survival has clearly emerged for other malignancies as well, such as kidney cancer [20]. It can be speculated that traditional progression-based endpoints are not ineffective per se, but it is the timing of their measurement that is inadequate [19]. Although there was no difference in time from enrolment to ‘first’ tumour progression in the IMPACT trial, on the hypothesis of a delayed effect of immunotherapy, subsequent tumour progressions (e.g. time from ‘first’ to ‘second’ tumour progression) could have shown benefit for patients enrolled in the active arm, provided that crossover and unblinding were not allowed and clinical management after progression had been homogeneous and defined in detail by the protocol for both arms – three conditions that were not implemented in the IMPACT trial. The validation of available biomarkers of immune activation (such as antibody titres measured by ELISA test or T-cell cytokine production measured by ELISPOT test), and the development of new ones may play a significant role for investigating immunotherapy mechanisms of action, assessing its efficacy and selecting patients in both clinical trials and clinical practice.

With a cost of US$ 93 000 per course of treatment, sipuleucel-T is surely one of the most expensive cancer treatments on the market. After its decision to review scientific evidence about sipuleucel-T, the Centers for Medicare and Medicaid Services concluded in March 2011 that it is a ‘reasonable and necessary’ therapy for prostate cancer [8], although the final decision to provide full coverage of sipuleucel-T was issued by the Centers for Medicare and Medicaid Services on 30 June 2011, from which time sipuleucel-T will be covered by the Medicare Program for men older than 65 years in the USA [21]. One opportunity missed by the Centers for Medicare and Medicaid Services review was to perform an economic evaluation of sipuleucel-T, which may have added valuable information to the Food and Drug Administration review. In a scenario where expensive and effective treatments are increasingly available, the inevitably limited financial resources mandate a formal evaluation of the economic burden that society can tolerate for the treatment of incurable diseases – a need that has become dramatically stringent with sipuleucel-T.

Sipuleucel-T has opened a new era for the treatment of prostate cancer. The incredible mixture of clinical and social issues related to this agent will alter clinical practice, direct future trials and pose questions and possible reforms to healthcare systems in the USA [8] and other Western countries.


None declared.