Results from a
Phase III trial that attempted to shorten the treatment duration for drug-sensitive
pulmonary TB from four to six months, by using gatifloxacin instead of
ethambutol, showed that the shortened treatment duration was inferior when compared to
the standard treatment course. Results from the
OFLOTUB trial were presented earlier this month at the 44th World Conference on Lung Health in Paris.
A previous study, RIFAQUIN, found that it was not possible to shorten the TB treatment regimen from 6 months to 4 months by use of moxifloxacin and rifapentine twice weekly in place of rifampicin and isonazid in the continuation phase of TB treatment.
The non-inferiority, blinded, randomised control trial
was conducted in Benin, Guinea, Senegal, Kenya, and South Africa. In the control arm of the study, 919 people were given the World Health Organization-recommended six-month
course of TB treatment, whereas 917 people in the test arm received 400mg of
gatifloxacin, six days a week, in place of ethambutol for two months, followed
by two months of treatment with gatifloxacin, rifampicin and isoniazid.
The median age for participants in the study was 31 years, 27% were female, and 18% had HIV. All participants were adults with new TB
diagnoses.
The study used a 6% non-inferiority margin, and
considered patients’ “unfavourable” outcomes as a composite endpoint at two
months, end of treatment, 18 months, and 24 months, with the primary outcome
being at 24 months. Researchers considered participants who died, were lost to follow-up during
treatment, experienced treatment failure, or experienced TB re-occurrence (a
category that included relapse and re-infection), as experiencing unfavourable
outcomes.
At two months and end of treatment, those in the
intervention arm fared better. However, 24 months after completion of treatment,
a larger proportion of people in the gatifloxacin-containing
arm had unfavourable outcomes (20.9 vs 16.8%). This difference was driven by a
higher frequency of tuberculosis recurrence in the gatifloxacin arm (14.6 vs 6.9%
of participants were diagnosed with TB again). The intent-to-treat analysis showed
a difference of 3.8% favouring the control arm, with a confidence interval that
exceeded the non-inferiority margin (-0.03%, +8%). As a consequence the
gatifloxacin-containing regimen was found to be inferior to standard TB
treatment.
Importantly, the gatifloxacin-containing regimen was
considered safe and well-tolerated: participants taking it were less likely to experience a
QT prolongation in the test arm than the control arm, with a risk difference
of -0.1% in the test arm (95% confidence interval -0.2, 0.1). "This
is very important because almost all the [new] drugs in the pipeline have QT
prolongation issues," says clinical investigator Corinna Merle, who
presented the findings. "When it comes to mixing new TB drugs, you might
not want to mix two drugs with QT prolongation." The risk of hyperglycemia was "almost the same in
both arms", she says.
Interestingly, different countries demonstrated
different outcomes: people fared better in the intervention arm in Benin,
Guinea, and Kenya than those in Senegal and South Africa. Noting
that countries where participants had lower BMIs when they began treatment – such as
Benin and Guinea – did better on treatment than those with higher initial
BMIs – such as Senegal and South Africa – Merle wonders whether the regimen's
effectiveness could be increased with better dosing. (Patients in Kenya and
Senegal, however, had similar BMIs at baseline, yet Kenyan patients did better
in the intervention arm than Senegalese patients.) All patients, regardless of
body weight, were given 400mg a day.
A pharmacokinetic
study is currently being conducted in South Africa to help determine
appropriate dosing. Results are expected within the next six months. Merle also
suggests that giving treatment every day of the week, instead of six days out of
seven, may be important when attempting to use a shortened regimen (the team
decided to treat six out of seven days as many participating countries did not
have healthcare centres open on Sunday). Participants
who were living with HIV also did better in the intervention arm than those who did not have HIV.
Merle says that
the intricacies found in the OFLOTUB trial suggest that more research should be
conducted, and that gatifloxacin shouldn't be discounted as a potential TB
regimen yet. "In the control arm we had
more patients lost to follow-up," Merle says. "And then more
treatment failure in the control arm, but what you have more of in the
[intervention arm] is re-occurrences.” Merle notes that poorer outcomes in the intervention
arm at 24 months may simply be because patients who had taken gatifloxacin were
followed-up for longer than those on standard treatment, as they were taking two
months' less treatment. “We need to look at the dynamics in more
detail to explain all of this, and why patients in some countries did better
than others. We also need to look more at dosing, and at gatifloxacin treatment
in HIV-positive patients."
Merle also notes
that gatifloxacin is a generic drug, whereas moxifloxicin – being considered for
use in a shortened TB treatment regimen as part of the REMox trial – is not. Given that
both OFLOTUB and REMox are attempting to shorten TB treatment periods, Merle
says "lets see [the REMox] results. If they have very good results,
there's no reason to have two types of regimens. But if they have results
similar to ours, then there could be a discussion…I really think that [gatifloxacin]
should not be buried…For the moment we will wait."