Granulocyte-colony-stimulating Factor for Acute Ischemic Stroke: A Randomized Controlled Trial (STEMTHER)

Abstract

Granulocyte-colony-stimulating factor (G-CSF) functions both as a neuroprotectant and a stimulator of autologous bone marrow stem cell release. Therefore, administration of G-CSF should improve the outcome of stroke. Here, we examine the safety of using G-CSF to treat acute ischemic stroke using a randomized controlled trial involving 20 adult patients presenting with ischemia in the carotid region within 48 h of onset. The experimental group (n = 10) received subcutaneous G-CSF injections (10 mg kg(-1) day(-1)) in addition to conventional therapy for 5 days. The primary outcome was motor function as measured by the modified Rankin Scale 180 days post-stroke. Safety was evaluated according to the frequency of hemorrhagic transformation of infarctions and serious adverse events. Only six patients in the experimental group completed full course of treatment, while four patients (three in the control and one in the experimental group) were lost to follow-up. We found the experimental and control groups did not differ significantly in either neurological impairment or degree of disability/dependence at 180 days post-stroke. We conclude that while adding G-CSF (10 mg kg(-1) day(-1)) to acute ischemic stroke therapy for 5 days is safe, its efficacy remains unproven.

The Efficacy and Safety of Granulocyte Colony-Stimulating Factor for Patients with Stroke

“…Only 4 studies 14,22,24,25 reported allocation concealment using sealed opaque envelopes. All trials were double blind to participants and outcome assessors except 3.…”

“…Meta-analysis of 6 studies [20][21][22][23]25,27 with 494 patients demonstrated an insignificant difference in SAE (RR, 1.12; 95% CI, .91-1.38; P 5 .28). Six studies 14,[19][20][21][22]27 with 450 patients showed nonsignificant differences in the death of SAE (RR, 1.25; 95% CI, .82-1.91; P 5 .30).…”

“…Six studies 14,[19][20][21][22]27 with 450 patients showed nonsignificant differences in the death of SAE (RR, 1.25; 95% CI, .82-1.91; P 5 .30). There were no significantly statistical differences in the rates of infection (RR, .87; 95% CI, .53-1.43; P 5 .59) and venous thromboembolism (RR, .83; 95% CI, .21-3.27; P 5 .79) between G-CSF and placebo.…”

“…The total leukocyte count also increased either in a fixeddose regimen or a dose-dependent manner. Seven studies 14,[19][20][21][22][23][24] with 514 patients showed that G-CSF had significantly elevated the leukocyte count than the placebo group (SMD, 3.52; 95% CI, 2.54-4.49; P , .001), but, which did not induce an increase in vascular complications.…”

“…31 Similarly, well-tolerated administration, feasible treatment regimen, and improved neurologic outcomes have been reported in the individual studies. 14,21 As to the clinical and functional outcomes, some studies 15,16,[20][21][22]27 found that the G-CSF did not show any progress in clinical outcomes at 90 days, although trends toward improvement were noted in change in the NIHSS 15 and BI scores. 21 In other randomized, double-blind studies, [23][24][25][26]32 they detected within 7 days of stroke onset and observed that there was a consistent trend toward improvement in neurologic functional recovery in the intervention group.…”