4. Discussion
Since vigabatrin was initially approved in the US in 2009, a total
of 1200 adult patients have enrolled in the Sabrilฎ registry, seizures in
whom had failed to respond to many other therapies and had rCPS. At
enrollment, the majority of adult patients had prior treatment with 4
to 9 AEDs; most were receiving 2 or more concomitant AEDs when
starting vigabatrin (and therefore, may be more likely to have seizures
that may be difficult to control). Though participation in the registry is
mandatory for all US patients prescribed vigabatrin, 35% of the adult patients
for whom a baseline OAF was submitted were exempted by their
physicians from vision testing; reasons included underlying comorbid
conditions (mainly neurologic disease including cognitive disability),
preexisting blindness, and underlying medical conditions.
Despite having an enriched proportion of patients with highly refractory
epilepsy, two-thirds of patients who enrolled at least 3 months
before the datacut continued vigabatrin for N3 months, and 42% of patients
who enrolled at least 12 months before the datacut continued in
the registry for over 12 months. This suggests that a large proportion
of patients with highly drug-resistant epilepsy and their physicians
felt that the patients had benefited from vigabatrin treatment. This is
notable given the requirement that patients continue to participate in
the Sabrilฎ registry to receive vigabatrin and the need to continue to
screen for visual risks with regular visual monitoring.
Retention was greater in the small group of patients who were
previously exposed to vigabatrin than patients “na๏ve” to vigabatrin,
likely because they represented a group of patients who had already
been established as “vigabatrin responders”.
The REMS for vigabatrin was designed to reduce the risk of
vigabatrin-induced vision loss by (among other measures) requiring
periodic benefit–risk assessments through vision monitoring and to
mitigate potential risks by discontinuing vigabatrin therapy for patients
who experience inadequate clinical response. Nearly all adult patients
had seizures that were highly refractory to alternative therapies and
thus, consistent with the approved indication. The registry documented
a gradual reduction in the proportion of patients continuing therapy
over 3 to 12 months; the reduction from 70% to 42% supports
physicians and patients performing appropriate benefit–risk assessments.
Although changes in seizures were not included in the registry,
this pattern suggests that a substantial proportion of patients remained
on therapy after the mandatory benefit–risk assessment at 3 months.
The Sabrilฎ registry is the largest prospective collection of patients receiving
vigabatrin ever described and accurately profiles demographic
data for adult patients with rCPS in the US.
Despite these strengths, the registry has several limitations. First,
data accuracy depends on the quality and completeness of reporting
by each prescriber. Some errors in recording likely occurred because
of a lack of standardized form completion and inaccurate data entry
by some clinicians. Examples include: (1) a small number of adult patients
in the registry were identified as having IS, a diagnosis typically
made in patients under 3 years of age (it is likely that adults with IS
are probably patients with IS early in life who had this diagnosis
coded by registry clinicians); (2) somepatientswere recorded as having
taken no AEDs prior to enrollment (given vigabatrin's indication and orphan
indicationwith REMS, it is unlikely that vigabatrinwould be given
as initial therapy in adults); (3) some patients were reported as having