Freed: sham-controlled fetal grafts

The research record

This is the project's file for this landmark, as written and checked. Sources are listed at the end; anything not yet confirmed is marked "Not yet verified".

Who

Curt Freed (University of Colorado School of Medicine, Denver, per the listed affiliation), Paul Greene, Robert Breeze, Stanley Fahn, David Eidelberg, John Trojanowski and colleagues (12 authors in all). Published in the New England Journal of Medicine, 8 March 2001 issue.

What happened

Open-label grafts reported from the late 1980s (landmark 3) had no placebo group. This trial added one.

• 40 patients aged 34 to 75, with severe Parkinson's (mean duration 14 years), were randomly assigned to a transplant or sham surgery. In the sham group, holes were drilled in the skull but the dura (the membrane over the brain) was not pierced.
• Transplant patients got cultured midbrain tissue from four embryos in the putamen (the main dopamine target area) on both sides.
• Main measure: patients' own global rating of change at one year, on a scale of -3 to +3. The mean was 0.0 (SD 2.1) with transplant and -0.4 (SD 1.7) with sham. No significant difference.
• In patients 60 or younger, morning pre-medication motor scores were better with transplant (UPDRS P = 0.01; Schwab and England P = 0.006). Older patients did not improve.
• Fibre outgrowth from the graft showed up on fluorodopa PET scans or at autopsy in 17 of 20 transplanted patients.
• After the first year, dystonia and dyskinesias (involuntary movements) returned in 15% of transplant recipients, even after levodopa was cut.

Why it moved us toward a cure

It showed that grafted fetal neurons survive and grow in the human putamen, in a controlled design that open-label reports could not give. It also exposed the problems that later work had to solve: a benefit limited to younger patients and side-effect movements caused by the graft itself (landmark 215). A second sham-controlled trial followed (landmark 318). The animal work behind it is in landmark 23.

Outcome

Dead end: the graft did not beat sham on the main measure. The approach was limited by tissue supply (four embryos per patient) and the graft-induced movements. Work moved to stem-cell-derived neurons (landmark 10).

Sources

• Freed et al., N Engl J Med 2001;344:710–719, DOI 10.1056/NEJM200103083441002 (PubMed 11236774; abstract read via Europe PMC)
• Not yet verified: the institutions of the authors other than the first; enrolment and surgery dates (a secondary source gives enrolment April 1994 to April 1997, which I could not confirm); the full text (NEJM is paywalled; every number above is from the abstract); whether the sham group later received transplants (a bioethics review says 14 accepted real surgery, which I did not trace to the primary paper); the funder.