SEATTLE — A substantial proportion of clinical trials involving neurologic diseases or interventions within the National Institutes of Health do not require pregnancy testing and do not offer advice or counseling on effective contraceptive methods to women of childbearing potential even though participation involves potential risks, according to a systematic review of intramural clinical trials at the institutes.
In the 15 years since the Food and Drug Administration formulated guidelines for the study and evaluation of gender differences in the clinical evaluation of drugs, “there have been no published studies on whether the protections for women are being implemented,” said Dr. Barbara Karp, chair of the Combined Neuroscience Institutional Review Board for six of the NIH institutes.
The guidelines withdrew the restriction on the participation of women of childbearing potential (WOCBP) and placed protections for them that included the use of appropriate contraception or precautions against a woman becoming pregnant, counseling on the importance of not becoming pregnant, testing to make sure that they are not pregnant on study entry, and making participants aware of the teratogenic risks associated with participation.
Dr. Karp and her colleagues evaluated the protocols and consent and assent forms for all active clinical trials in the NIH Intramural Research Program as of June 2008 that included neurologic illnesses primarily affecting the central or peripheral nervous system and neurologic interventions (brain or spinal cord imaging with CT, PET, or MRI; brain stimulation with transcranial magnetic stimulation or transcranial direct-current stimulation; or neurosurgery).
Of 583 active clinical trials with protocols that were studying neurological interventions or neurological diseases with stages that could affect women during the ages of menarche through menopause, the investigators identified 84 (14%) with WOCBP.
All of the trials excluded pregnant women. Most (93%) of the trials involved drugs in FDA classes B, C, D, or X. About one-third of the trials exposed women to radiation, mostly from PET scans but also from CT scans and fluoroscopy. MRI scans, which have never been shown to be safe to a fetus and were therefore deemed indeterminate in safety, occurred in 53% of the trials. Another 7% used brain stimulation.
“We found that even though pregnant women were excluded from the trials, the nature of the exact type of risk and the specific reason for the exclusion was not always identified in the protocol,” Dr. Karp said at the annual meeting of the American Academy of Neurology.
Drug risks were mentioned most often in protocols, whereas radiation risks were rarely acknowledged.
MRI risks were rarely conveyed. However, possible risks of transcranial magnetic stimulation were mentioned.
On consent forms, only 48% of all the risks posed to a fetus were specified.
In some trials, there were inconsistencies between protocols and consent forms in the descriptions of the natures of risks. Protocols were more likely to describe the risks posed by drugs than were consent forms, whereas consent forms were more likely to mention radiation and MRI as posing a risk to a fetus.
In 18 studies with interventions that involved only episodic risk to a fetus (duration of exposure was one menstrual cycle or less), 94% tested for pregnancy at study entry and 67% repeated testing immediately before an intervention that posed risk.
In 66 studies that involved a continuous risk of exposure (more than 1 month or an indeterminate duration), 92% tested for pregnancy at study entry and 47% repeated the testing during the study. In studies that did not perform testing, patients were asked about possible pregnancy at visits and were withdrawn if they were pregnant.
Only 2% of the protocols of trials that involved episodic risk required contraception and none of them listed effective contraceptive methods. However, 75% of the consent forms for trials involving episodic risk required contraception.
In the study protocols that posed continuous risk, 68% required contraception and 42% listed effective contraceptive methods. Similarly, 69% of consent forms for studies with continuous risk required contraception and 42% listed effective methods.
“We think that listing effective methods is especially important in studies with drugs that can interfere with oral contraceptives or make them less effective because women won't know otherwise that they're actually using a form of birth control that might not work when they are participating in the study,” said Dr. Karp, who said that she had no relevant disclosures.
“Consents and protocols were generally silent on whether contraceptive counseling was offered and since about half of pregnancies in the United States are unintended—usually the cost of improper or inconsistent use of contraceptives—providing information on contraceptive methods and instruction on proper use is important.”
In six studies that enrolled participants younger than 18 years and involved continuous risk, the minor assent forms of five of the studies required pregnancy testing, but only three informed subjects of the need. The protocols of three of these studies required contraception, but only one study told participants of this requirement.
FULL TEXT