STEMMI Trial data indicate that noninvasive injections of the stem cell mobilization factor granulocyte-colony stimulating factor (G-CSF) are no better than placebo at repairing myocardium damaged by infarction, according to a presentation at the annual meeting of the American College of Cardiology.

Based on earlier animal and human research, researchers hypothesized that G-CSF would increase release of adult stem cells from bone marrow into the bloodstream. Stem cells that entered the myocardium were expected to repair damaged myocardium by differentiating into new muscle cells.

Patients in the study had an ST-elevation myocardial infarction (STEMI), the type that is typically treated by balloon angioplasty and stenting. However, as Kastrup remarked, opening previously blocked arteries does not repair damaged myocardium.

Several small clinical studies have suggested that infusing adult stem cells through a catheter after stenting improved the heart function of some STEMI patients. However, the technique is invasive, and the studies were neither double-blinded nor placebo-controlled, leaving the validity of the results in question. Recent experiments have also indicated that G-CSF may have a direct effect on damaged but viable myocardium.

In the randomized, double-blind, placebo-controlled trial called STEM cells in Myocardial Infarction (STEMMI), patients underwent angioplasty and stenting at University Hospital Rigshospitalet in Copenhagen less than 12 hours after onset of symptoms.

The 62 men and 16 women (average age 56 years) were evenly randomized to receive one daily subcutaneous injection of G-CSF or placebo for six days. Kastrup and colleagues began the injections 10 to 65 hours after stenting, and 85 percent of patients were treated within 48 hours.

G-CSF proved safe in the patients. However, six months after injections, researchers found no evidence that the stem cell mobilization factor had provided any significant benefit. The study indicated that patients injected with G-CSF had a 17-percent increase in myocardial thickness around their infarction site. However, myocardial improvement immediately surrounding scar site (border zone) and in non-infarcted myocardium tended to be lower in the G-CSF group. The placebo group showed increases of 23 percent and
19 percent for border zone and non-infarcted areas, respectively. Those differences are explained by differences that existed between the two groups at enrollment.

Notably, there were no significant differences between the two patient groups in left ventricular mass or post-systole left ventricular blood volume at six months. There was also no change in the size of dead myocardium at six months, despite similar sizes at enrollment.

"Our trial only settles the question regarding G-CSF injections initiated soon after an acute heart attack and using one specific dose without accompanying angiogenic therapy to increase blood vessel growth," said senior author and presenter Jens Kastrup, MD, associate professor of cardiology at the University of Copenhagen in Denmark. "We did not test another dose or another time, such as three weeks after an attack when some angiogenic factors in the blood have reached peak concentrations."

The negative findings may be due to small sample size, but Kastrup
acknowledged that these results are not encouraging.

"A significant finding is a prominent improvement in the cardiac function of the placebo group," he said. "Due to this result, we strongly recommend a double-blind, placebo-controlled design in any future studies of G-CSF after a heart attack."