Ann: DMC Completes Final Review of Phase 3 Heart Failure Trial, page-45

Basically they changed the protocol after the study had started to allow consideration of these subsequent (previously censored) events which is a bit of a no-no in general but I think in this case will be accepted by readers.

In this particular space the FDA are the final arbiters of the rules, for better or for worse, readers don't count, they just go along for the ride .

"Extensive discussions were conducted with the Food and Drug Administration (FDA) on modifying the study’s primary efficacy end point from a TTFE composite to one using recurrent events with the analysis based on the joint frailty model (JFM)."

"This was accomplished using an FDA-reviewed protocol amendment and revised statistical analysis plan that allowed for adaptive patient enrichment of baseline NYHA class III patients during the screening evaluation."

Some additional requirements were put in place with the change to the trial design

"It is important and required by FDA to verify that improvement in recurrent HF hospitalizations is not at the expense of worsening TCEs. Accordingly, the following prespecified analyses will be performed to assess the beneficial active treatment effects that decrease the hazard ratio relative to controls:

• delayed or less frequent nonfatal HF-MACE (primary end point: analysis based on the JFM) and
• nonincreased, delayed, or lower TCE rate (key secondary end point: analysis based on TTFE)."

But the stem cell patient is admitted to hospital in heart failure 6 times in the next year before returning to running marathons. Does that mean the stem cell treatment improves survival (yes) or does it actually increases hospital admissions with heart failure (yes because it increases survivors).

The primary end point in the trial was the reduction in HFMACE, the time to first TCE was a secondary endpoint, as were related in or outpatient visits. So a patient admitted to hospital for any HFMACE 6 times in the next year before returning to running marathons would contribute to failure from the primary endpoint view if they were in the treatment arm, and they would probably contribute to trial failure if they progressed to a TCE too.

The aim was to reduce both HFMACE but not at the expense of increased TCEs (and other lesser issues based on the secondary endpoints).

I did overstate the effect of all the Terminal Cardiac Events, a few may be fortunate;

"The TCEs were defined as cardiac death, LVAD implantation, heart transplant, or placement of an artificial heart. Only the first TCE was taken into account in the primary and key secondary analysis."

Finally I am having trouble reconciling your comment;

This is what they called censoring, the later events are censored.
Basically they changed the protocol after the study had started to allow consideration of these subsequent (previously censored) events....

With this from the review, bolded for emphasis;

"This significant excess of TCEs in the more vulnerable class III HF patients and their subsequent censoring of recurrent nonfatal HF-MACE from the trial are critical factors because the TCE-censored class III patients had a significantly higher rate of primary end point events before the time of censoring than did the TCE-censored class II patients. Over time, the expected result of disproportionately losing class III patients would be an important reduction in the number, as well as prolongation of the timing of future recurrent nonfatal HF-MACE. This is an example of how blinded data acquired during the conduct of a clinical events-driven trial can indicate an operational design problem that was not anticipated at the start of the study."

That implies to me the censoring is of earlier events - before the time of censoring, i.e. if a patient has 3 HFMACE then goes on to record a TCE the HFMACE are censored, is that an incorrect interpretation? I have read that censoring is supposed to apply to events subsequent to a defined event but that does not seem to be the case here and if it were applied post event would seem to negate the need for enrichment, unless these large number of HFMACE were happening post TCE which to this layperson seems nonsensical given the events defined as a TCE.

My conundrum is further compounded by this from the review;
"Only the first TCE was taken into account in the primary and key secondary analysis"