According to the Clinical Trials Registry (https://clinicaltrials.gov/ct2/history/NCT04614948?V_14=View#StudyPageTop) ENSEMBLE 2 is expecting completion in May, 2022. The study design is also two doses of placebo vs. two doses of Ad26.COV2.S (not 1 vs. 2 doses as I personally would have expected). The two-placebo subjects would receive a single dose of the vaccine at unblinding, but that page doesn't state when unblinding would occur.
I am guessing you were expecting preliminary data by now?
This is the thing about ClinicalTrials.gov; it's not very approachable. The study completion date isn't always the date that the "final analysis" is actually released. Often, particularly in studies with long safety follow-up or that use a surrogate endpoint to predict clinical outcomes, there is a projected study duration that represents the expected completion date of the trial, but doesn't represent the data that people are most interested in.
In this case, the safety evaluations for most of the COVID-19 vaccines are expected to continue for about two years past study onset, which is probably was generated this date.
The efficacy evaluation, meanwhile, is event-driven, and so won't have a fixed final analysis date. Instead, it will depend on when enough events have accrued in the trial (events here being COVID-19 cases). Since there was no shortage of COVID-19 events over the winter when the trial was expected to read out, I'm really confused as to why it hasn't read out yet.
I have to admit that I don't love that trial design. I think it would have been better to do 3 arms, with 1 dose of J&J and 1 dose of placebo vs 2 doses of J&J vs 2 doses of placebo. But that would have taken even longer to enroll. It was probably for expedience that they designed it this way--doing 1 dose vs 2 doses might have been less informative about the 2-dose regimen if the numbers had been very close. This gets a cleaner result that stands to be more cross-comparable to other trials (even though you shouldn't cross-compare trials).
Not to make this comment thread too long, but ... given that circumstances inevitably change over long trials (different strains of the pathogen, different incidence in the population, different levels of background immunity) surely it's especially hard to compare trials that don't overlap in time?
Yes, extremely. Particularly when the disease is infectious. Other factors that impact cross-comparability are patient demographics, differences geographically in behavior and disease control measures, knowledge of how to treat or prevent the disease, etc.
According to the Clinical Trials Registry (https://clinicaltrials.gov/ct2/history/NCT04614948?V_14=View#StudyPageTop) ENSEMBLE 2 is expecting completion in May, 2022. The study design is also two doses of placebo vs. two doses of Ad26.COV2.S (not 1 vs. 2 doses as I personally would have expected). The two-placebo subjects would receive a single dose of the vaccine at unblinding, but that page doesn't state when unblinding would occur.
I am guessing you were expecting preliminary data by now?
This is the thing about ClinicalTrials.gov; it's not very approachable. The study completion date isn't always the date that the "final analysis" is actually released. Often, particularly in studies with long safety follow-up or that use a surrogate endpoint to predict clinical outcomes, there is a projected study duration that represents the expected completion date of the trial, but doesn't represent the data that people are most interested in.
In this case, the safety evaluations for most of the COVID-19 vaccines are expected to continue for about two years past study onset, which is probably was generated this date.
The efficacy evaluation, meanwhile, is event-driven, and so won't have a fixed final analysis date. Instead, it will depend on when enough events have accrued in the trial (events here being COVID-19 cases). Since there was no shortage of COVID-19 events over the winter when the trial was expected to read out, I'm really confused as to why it hasn't read out yet.
OK, but why test 2 doses vs. pure placebo, instead of 1 dose? If only for ethical reasons?
I have to admit that I don't love that trial design. I think it would have been better to do 3 arms, with 1 dose of J&J and 1 dose of placebo vs 2 doses of J&J vs 2 doses of placebo. But that would have taken even longer to enroll. It was probably for expedience that they designed it this way--doing 1 dose vs 2 doses might have been less informative about the 2-dose regimen if the numbers had been very close. This gets a cleaner result that stands to be more cross-comparable to other trials (even though you shouldn't cross-compare trials).
Not to make this comment thread too long, but ... given that circumstances inevitably change over long trials (different strains of the pathogen, different incidence in the population, different levels of background immunity) surely it's especially hard to compare trials that don't overlap in time?
Yes, extremely. Particularly when the disease is infectious. Other factors that impact cross-comparability are patient demographics, differences geographically in behavior and disease control measures, knowledge of how to treat or prevent the disease, etc.