Sipuleucel-T (1) – Provenge®

• Clinical trials
  • To assess the additional benefit of sipuleucel-T, the pharmaceutical manufacturer submitted three relevant studies: IMPACT, D9901 and D9902A, which are randomised, double-blind, multicentre registration trials comparing Sipuleucel-T with a placebo.

asymptomatic or minimally symptomatic, metastatic (non-visceral), castration-resistant prostate cancer in adult men for whom chemotherapy is not yet clinically indicated

• Hint for a non-quantifiable additional benefit
• The certainty of the evidence (probability of additional benefit) is classified as ‘hint’.
• There is uncertainty regarding the validity of the results.
• Mortality – overall survival
  • The results for the endpoint of overall survival can only be interpreted with limitations for the benefit assessment.
  • The dossier described a prolongation of the median overall survival of 4.1 months in the IMPACT study. (Median time to event in the sipuleucel-T arm was 25.8 months [95% CI 22.8; 27.7] and 21.7 months [95% CI 17.7; 23.8] in the control arm.)
  • The studies D9902A and D9901 also showed an advantage in overall survival in the sipuleucel-T group.
  • Due to the differences described between the treatment arms and the lack of information on subsequent therapies, these figures cannot be interpreted with the necessary validity.
  • As the influence of concomitant treatments on the effect estimates for all studies cannot be definitively ruled out, the extent of the effect cannot be conclusively assessed and is therefore non-quantifiable.
• Morbidity – time to onset of disease-related pain
  • For the endpoint ‘time to onset of disease-related pain’, data were available for asymptomatic patients in the three studies IMPACT, D9901 and D9902A.
  • The meta-analysis shows no statistically significant difference between the treatment arms.
  • An additional benefit of sipuleucel-T compared with the appropriate comparator therapy (watchful waiting) is therefore not proven for this endpoint.
• Morbidity – PFS
  • The study documents for all three trials indicate that no statistically significant difference in disease progression was observed between the two treatment arms.
  • Median time to progression (in months, intervention vs. control): 3.4 vs. 3.3 (IMPACT), 2.7 vs. 2.3 (D9901), 2.5 vs. 2.3 (D9902A). Hazard ratios (with 95% confidence intervals): 0.94 (0.77; 1.15) (IMPACT); 1.45 (0.99; 2.11) (D9901); 1.09 (0.69; 1.70) (D9902A).
  • An additional benefit of sipuleucel-T compared with the appropriate comparator therapy (watchful waiting) is therefore not proven for this endpoint.
• Health-related quality of life
  • Health-related quality of life is a patient-relevant endpoint in the context of benefit assessment.
  • Data on health-related quality of life were not collected in the three studies.
  • There is therefore no proof of additional benefit from sipuleucel-T compared with the appropriate comparator therapy (watchful waiting) for this endpoint.
• Side effects
  • Leukapheresis-related side effects (i.e. AEs occurring within one day of leukapheresis) occurred equally in both treatment groups across all three studies.
  • In the IMPACT study, approximately 20% of patients in both groups experienced citrate toxicity.
  • These side effects would not have occurred under a watch-and-wait approach.
  • The harm associated with sipuleucel-T compared with the appropriate comparator therapy (watchful waiting) may be underestimated in relation to these events.
  • For the SAE endpoint, the meta-analysis of the three studies showed no statistically significant difference between the treatment arms.
  • There is therefore no proof of greater or minor harm from sipuleucel-T compared with the appropriate comparator therapy (watchful waiting) for this endpoint.
  • For the endpoint of severe AEs (CTCAE grade ≥ 3), the meta-analysis of the three studies showed no statistically significant difference between the respective treatment arms.
  • There is therefore no proof that sipuleucel-T causes greater or minor harm compared with the appropriate comparator therapy (watchful waiting) for this endpoint.
  • Results on discontinuations due to AEs were available only for the IMPACT trial, as, according to the principal investigator, data collection in the other two trials was unsystematic.
  • This study shows no statistically significant difference between the treatment arms.
  • There is therefore no proof that Sipuleucel-T causes greater or minor harm compared with the appropriate comparator therapy (watchful waiting) for these endpoints.
  • For the endpoint of fever, the meta-analysis revealed significant heterogeneity.
  • In all three studies, a statistically significant difference was observed between the treatment groups to the detriment of sipuleucel-T; the effects were thus clearly consistent.
  • For this endpoint, there is therefore an indication of greater harm from Sipuleucel-T compared with the appropriate comparator therapy (watchful waiting).
  • In the meta-analysis of the three studies IMPACT, D9901 and D9902A, a statistically significant difference was observed for the endpoints of headache and chills, to the detriment of sipuleucel-T.
  • For these endpoints, there is therefore an indication of greater harm from sipuleucel-T compared with the appropriate comparator therapy (watchful waiting).
  • The results regarding fever, headache and chills provide an indication that sipuleucel-T causes greater harm compared with the appropriate comparator therapy (watchful waiting).
  • On balance, this does not lead to a downgrading of the additional benefit, as the symptoms of fever, headache and chills are generally not serious and are easily treatable.
• Conclusion
  • For patients with asymptomatic or minimally symptomatic, metastatic (non-visceral), castration-resistant prostate cancer, for whom chemotherapy is not yet clinically indicated, an unquantifiable added benefit for sipuleucel-T compared with a watch-and-wait approach can be inferred from an overall assessment of the available results on mortality, morbidity and side effects, there is a non-quantifiable additional benefit for sipuleucel-T compared with a watch-and-wait approach.
  • Due to uncertainties in the study design, the extent of the additional benefit in terms of the mortality endpoint cannot be quantified.

Courtesy translation only, please refer to the German original.