Pregabalin, capsule, 25 mg, 75 mg, 150 mg and 300 mg, Lyrica® - March 2011
Page last updated: 04 July 2011
PDF printable version to Pregabalin, capsule, 25 mg, 75 mg, 150 mg and 300 mg, Lyrica® - PDF 88 KB
Public Summary Document
Product: Pregabalin, capsule, 25 mg, 75 mg, 150 mg and 300 mg, Lyrica®
Sponsor: Pfizer Australia Pty Ltd
Date of PBAC Consideration: March 2011
1. Purpose of Application
The submission sought an Authority Required (STREAMLINED) listing for the initial and continuing treatment of neuropathic pain in patients who meet certain criteria.
2. Background
This drug had not previously been considered by the PBAC for this indication.
3. Registration Status
Pregabalin was TGA registered on 13 April 2005 for the treatment of neuropathic pain in adults and as adjunctive therapy in adults with partial seizures with or without secondary generalisation.
4. Listing Requested and PBAC’s View
Note: Pregabalin is not subsidised for fibromyalgia or acute pain or chronic pain of non-neuropathic origin.
Authority Required (STREAMLINED)
Initiation and up-titration of treatment for neuropathic pain.
Authority Required (STREAMLINED)
Continuation of treatment in patients who have received a PBS prescription for initiation
of treatment and have shown an adequate clinical response.
Continuation of treatment in patients who had shown an adequate clinical response
to pregabalin prior to PBS-listing.
For PBAC’s view, see Recommendation and Reasons.
5. Clinical Place for the Proposed Therapy
Neuropathic pain refers to a specific pain syndrome characterised by pain and sensory
abnormalities in body parts that have lost their normal peripheral innervation or
sensory representation in the central nervous system. There are many types of neuropathic
pain, with classification based on the underlying condition eg. diabetic peripheral
neuropathy, postherpetic neuralgia, human immunodeficiency virus (HIV) neuropathy.
Current therapies used to treat neuropathic pain include tricyclic antidepressants,
gabapentin, carbamazepine, other antiepileptics, simple analgesics, opioid analgesics,
non-steroidal anti-inflammatory drugs (NSAIDs), selective serotonin re-uptake inhibitors
(SSRIs), selective noradrenaline re-uptake inhibitors (SNRIs), and complementary medicines.
Patients cycle between treatments, switching and adding other neuropathic pain agents.
Non-pharmacological treatments may also be useful in reducing and managing the pain.
The submission proposed that pregabalin would provide an additional PBS-subsidised
option for the treatment of neuropathic pain.
6. Comparator
The submission nominated amitriptyline as the main comparator; carbamazepine, gabapentin
and opioid analgesics are secondary comparators.
The PBAC considered the clinical comparison against both an active comparator (as
represented by the sponsor’s nominated main active comparator, amitriptyline) and
placebo to be reasonable.
For PBAC’s view, see Recommendation and Reasons.
7. Clinical Trials & Results of Trials
The clinical trials were presented by type of neuropathic pain - diabetic peripheral
neuropathy (DPN), post-herpetic neuralgia (PHN) and other.
DIABETIC PERIPHERAL NEUROPATHY (DPN)
Pregabalin versus amitriptyline:
The submission presented two trials (Trial 040 and Bansal 2009) which compared amitriptyline
and pregabalin. Trial 040 also included a placebo arm and no difference was shown
between pregabalin and placebo in change in mean endpoint pain score, proportion of
responders or mean Patient Global Impression of Change (PGIC). Amitriptyline was statistically
significantly better than placebo on all three measures. Both active treatments showed
quality of life improvements (SF-36) compared to placebo.
Bansal 2009 was a crossover study showing no differences between pregabalin and amitriptyline
on pain scores, responder rates or PGIC.
Pregabalin versus placebo:
The submission identified 12 additional trials comparing pregabalin to placebo, as well as trial 040. The submission performed stratified meta-analyses by dose of pregabalin. There were no statistically significant differences between pregabalin 75 mg, 150 mg/day and placebo in changes in pain scores. Pregabalin ? 300 mg/day was more effective than placebo (change in pain scores, responder rates, PGIC scores).
Amitriptyline versus placebo:
The submission identified two additional cross-over trials comparing amitriptyline
with non-active comparator. Only one eligible trial was identified for indirect comparisons
(Trial 040). There were statistically significantly larger reductions in pain scores,
more responders and improved PGIC scores in amitriptyline users compared to placebo.
Indirect comparison pregabalin vs amitriptyline (placebo common comparator):
The submission also presented meta-analyses of the results of 13 trials comparing
pregabalin with placebo at differing dosages and based on mean changes in pain scores,
responder analyses (? 50% reduction in pain scores) and Patient Global Impression
of Change (PGIC). The results of the placebo-controlled trials were used in indirect
analyses, comparing pooled estimates from the pregabalin trials with the amitriptyline
arm of Trial 040. There were no statistically significant differences between pregabalin
and amitriptyline 75 mg/day for any dose of pregabalin. Amitriptyline 75 mg/day was
associated with larger numerical reductions in pain scores than pregabalin for all
doses less than pregabalin 600 mg/day. Similarly, there were no statistically significant
differences in responder rates or mean PGIC scores between pregabalin and amitriptyline.
Numerically the estimates favoured amitriptyline. There were substantial placebo response
rates in the trials for DPN.
POST-HERPETIC NEURALGIA (PHN):
Pregabalin versus placebo:
Nine trials were identified, and stratified meta-analyses by dose of pregabalin were conducted. There were no statistically significant differences between pregabalin 75 mg and placebo in changes in pain scores. Pregabalin ? 150 mg/day was more effective than placebo (change in pain scores, responder rates, PGIC scores). Placebo response appeared generally lower in these trials in PHN.
Amitriptyline versus placebo:
Four trials were identified, three crossover studies were excluded. There were statistically
significantly larger reductions in pain scores (VAS) in amitriptyline users compared
to placebo (Graff-Radford 2000).
“OTHER” NEUROPATHIC PAIN:
Pregabalin versus placebo:
The submission presented five RCTs of pregabalin versus placebo (spinal cord injury, central post-stroke pain, post-traumatic peripheral neuropathic pain, HIV neuropathy and central neuropathic pain).
Amitriptyline versus placebo:
Seven RCTs compared amitriptyline with non-active comparators (spinal cord injury, post-mastectomy, chemotherapy-induced neuropathic symptoms, HIV neuropathy, and amputation-related pain).
Indirect comparison pregabalin vs amitriptyline (placebo common comparator):
The submission pooled the results of four pregabalin versus placebo trials and two
amitriptyline trials. The pooled results for pregabalin and amitriptyline trials in
“other” neuropathic pain conditions and the individual trials for spinal cord injury
were used in indirect comparisons. For ‘other’ neuropathic pain conditions, pregabalin
150-600 mg/day was numerically superior to amitriptyline 10-125 mg/day, but there
were no statistically significant differences. However, for spinal cord injury there
was a statistically significant difference favouring amitriptyline.
OTHER COMPARATOR THERAPIESTop of page
Carbamazepine:
The submission included one systematic review, two RCTs in diabetic peripheral neuropathy
(DPN), and four RCTs in trigeminal neuralgia comparing carbamazepine versus placebo.
In DPN, two crossover trials were identified, no treatment comparisons were possible.
A Cochrane review (Wiffen 2005) concluded carbamazepine provides better pain relief
than placebo in several neuropathic pain conditions including DPN.
No trials were identified in PHN. A Cochrane review (Wiffen 2005) concluded that carbamazepine
provides better pain relief than placebo in several neuropathic pain conditions including
PHN.
Gabapentin:
The submission included seven RCTs in DPN, three RCTs in post-herpetic neuralgia (PHN),
one RCT in a mixed population (DPN and PHN), and 15 RCTs in “other” neuropathic pain
conditions comparing gabapentin versus placebo.
Of the seven trials identified in DPN, two cross-over studies were excluded. The trial
data suggested gabapentin 3600 mg/day is statistically significantly better than placebo
in DPN.
Three trials were conducted in PHN, one trial was conducted in a mixed PHN/DPN population
(this was excluded as it was a crossover trial). No meta-analysis was possible due
to differences in doses of gabapentin used in the trials. The individual trial results
suggested that gabapentin (starting dose 900 mg/day) is statistically significantly
better than placebo in PHN.
In “other” neuropathic pain, no individual study showed significant improvements of
pain intensity in the gabapentin vs placebo comparison. Only two studies were meta-analysed
for other/mixed types of neuropathic pain, and no significant improvement was demonstrated
in this comparison.
Opioids:
The submission included six RCTs in DPN, three RCTs in PHN, one RCT in a mixed population
(DPN and PHN), and ten RCTs in “other” neuropathic pain conditions comparing opioids
with placebo/non-active comparator.
Of the six trials conducted in DPN, one trial was in a mixed DPN/PHN population. Two
crossover studies were excluded. There were no common outcomes or dosing regimens
therefore no meta-analysis of trial results was possible. Tramadol and oxycodone CR
were associated with statistically significantly larger reductions in pain scores
than placebo. Tramadol was associated with larger proportions of treatment responders
and more improved PGIC scores than placebo.
Three trials were conducted in PHN, one trial involved a mixed PHN/DPN population.
Three crossover studies were excluded. A single parallel arm trial (Boureau 2003)
showed tramadol 100-400 mg/day (mean 275.5 mg/day) statistically significantly superior
to placebo in reductions in pain scores and was associated with significantly higher
responder rates than placebo.
In other/mixed types of neuropathic pain, only one study (Arbaiza 2007) in neuropathic
cancer pain has reported the primary meta-analysis outcome, the re-analysis result
did not show a significant pain score change from baseline with opioid (Tramadol).
For PBAC’s comments on these results, see Recommendation and Reasons.
8. Clinical Claim
The submission claimed pregabalin as non-inferior in terms of comparative effectiveness
and non-inferior in terms of comparative safety over amitriptyline, which the PBAC
considered acceptable.
For PBAC’s view, see Recommendation and Reasons.
9. Economic Analysis
A modelled economic evaluation was presented, comparing two scenarios: ‘Current Options’
representing current prescribing patterns of neuropathic pain drugs in Australia;
and ‘New Options’ based on expected prescribing patterns if pregabalin were listed
on the PBS for neuropathic pain. The model was a Markov model, over five years (10
six-monthly cycles).
Rather than moving patients between neuropathic pain health states, the model moved
patients between neuropathic pain treatment options, or treatment states. Each scenario
had five treatment states: anti-epileptic drugs (AED); tricyclic antidepressants (TCA);
tricyclic antidepressants plus anti-epileptic drugs (TCA+AED); other drugs; and no
drug treatment.
The treatment alternatives TCA, AED and TCA+AED were assumed to have the same efficacy.
Utilities, based on pain scores measured during the pregabalin trials were used to
derive QALYs. The costs included in the model were drug costs only. The only difference
between the three sub-models was in utilities. There were no differences in the costs,
or transition probabilities. The utilities used in the model are the same for each
arm of the model (‘Current option’ vs ‘New option’).
The transition probabilities for the ‘current options’ (pregabalin not on the PBS)
scenario were based on data from the sponsor’s Community Study. The Community Study
is a longitudinal patient audit of GP- and specialist-treatment of neuropathic pain
in Australia.
To derive transition probabilities in the ‘new options’ (pregabalin on the PBS) scenario,
the transition probabilities in the ‘current options’ arm are adjusted by multiplication
factors to reflect the expected increased use of pregabalin, decreased use of other
neuropathic pain drugs, and decreased numbers of patients receiving no drug treatment.
Based on the economic evaluation by type of neuropathic pain, an incremental cost
per QALY gained was calculated to be less than $15,000.
The results of the sensitivity analyses indicated that the model was most sensitive
to assumptions regarding utilities; the inclusion of a placebo effect among subjects
not taking drugs; and the assumed mean daily dose of drugs.
For PBAC’s view, see Recommendation and Reasons.
10. Estimated PBS Usage and Financial Implications
The financial cost per year to the PBS was estimated to be more than $100 million in Year 5.
11. Recommendation and Reasons
The PBAC accepted that there is a high clinical need for effective treatments for
neuropathic pain. Given the high unmet need, the requested restriction for a first-line
‘streamlined’ authority listing was considered appropriate. A listing that required
failure of other treatments was not considered practical for administrative reasons.
The PBAC considered the clinical comparison against both an active comparator (as
represented by the sponsor’s nominated main active comparator, amitriptyline) and
placebo to be reasonable. However, it was also considered probable that pregabalin
would be used in combination with other therapies such as amitriptyline, in which
case the comparator would be placebo, or other non-PBS listed anti-epileptic drugs,
such as gabapentin.
There were a number of issues with the clinical comparison, although the PBAC acknowledged
the claim in the submission that the head-to-head trial 040, which included amitriptyline
as an active control, lacked the power to show superiority or non-inferiority of pregabalin
versus amitriptyline. In addition, concern was raised that the results of the indirect
comparisons may not be particularly informative as the exchangeability of the trials
included in these analyses is uncertain due to the variability in the placebo responses
and heterogeneity between the trials. However, the PBAC acknowledged the Pre-PBAC
Response comments that the heterogeneity between the trials was accounted for by random-effects
meta-analysis.
Further uncertainty arises because of the doses used, and the short duration of the
fixed/maintenance doses, in the trials. Up to 100 mg per day of amitriptyline can
be given, and therefore there may have been underdosing of amitriptyline in Trial
040 (75 mg per day) and Bansal 2009 (maximum of 50 mg per day). The maximal dose for
pregabalin of 600 mg per day was used in these trials.
Nevertheless, the Committee concluded that despite the uncertainty about the comparative
effectiveness, the claim that pregabalin is no worse than amitriptyline, was acceptable,
despite the limitations in the data. The PBAC further noted that the submission made
no claim of therapeutic superiority for pregabalin over any other anti-epileptic or
tricyclic anti-depressant drug therapy for neuropathic pain.
The PBAC noted that the modelled economic evaluation compares two scenarios: ‘Current
Options’ representing current prescribing patterns of neuropathic pain drugs in Australia;
and ‘New Options’ based on expected prescribing patterns if pregabalin were listed
on the PBS for neuropathic pain. Thus, the model is effectively a comparison of pregabalin
with a range of alternatives, with the alternative strategy most widely being replaced
being no drug therapy. The PBAC noted that the modelled economic evaluation adopted
a broader approach to possible substitutions than generally promoted by the concept
of a “main comparator” in the PBAC Guidelines and considered it not to be unreasonable
under the circumstances, if substantial uptake beyond switching between currently
available drug therapies is expected. However, the PBAC was concerned whether the
model had addressed all possible alternative scenarios to the ‘Current options’ scenario,
for example the option of using drugs in sequence because pregabalin would most likely
be used early in the treatment sequence of neuropathic pain.
The PBAC noted that the economic evaluation is primarily driven by data from the Community
Study and assumptions about the neuropathic pain market size and market-share changes
if pregabalin is listed on the PBS. The Pre-PBAC Response argues that the Community
Study on which the transition probabilities were based, was a large, local and highly
relevant study. However, the submission did not provide a comprehensive study report
of the Study, but presented a large series of tables detailing the overall results
of the total population. The PBAC expressed concern about the lack of transparency
associated with presentation of the results of the Study. The limitations of the Community
Study included the small number of patients informing the transition probabilities
particularly in later cycles, the quality of information is subject to the adequacy
of record keeping, the eligibility to be selected is uncertain, and inconsistencies
in reporting of data. In addition, the submission acknowledged that it was not possible
to validate the results of the Community Study with the GPRN and BEACH surveys commissioned
by the sponsor due to different sampling and collection methods.
The approach of modelling a utility gain in patients who are not currently adequately
treated was not convincingly supported, and PBAC noted that this utility gain was
an important driver in the model due to the modelling effect of increasing the number
of patients moving to adequate drug treatment from inadequate drug treatment. The
submission also did not present the analyses of the individual patient data on utilities
for verification. Durations of treatments in the Community Study are not known. Furthermore,
the base case did not include placebo-adjusted responses.
The model assumes that all patients currently untreated and 38% of patients receiving
ineffective ‘other drugs’, have utility equal to the baseline utility of patients
in the reported clinical trials. This assumption is not supported by any evidence.
The PBAC also noted that there are inconsistencies in the mapping of changes in pain
scores to changes in utilities, such that a larger reduction in pain score seen in
DPN subjects (compared with PHN and other subjects) is associated with a lower utility
gain. Although the Pre-PBAC Response argued that trial-based utility data were used
in the model, the PBAC agreed with the ESC that the utility data were not convincing.
The PBAC thus considered that considerable uncertainty remained about both the transition
probabilities derived from the Community Study and utilities used in the model.
Given the high prevalence of the condition, the uptake of pregabalin would likely
be high, and possibly higher than that predicted in the submission.
The PBAC therefore rejected the submission because of uncertain cost effectiveness.
Recommendation:
Reject
12. Context for Decision
The PBAC helps decide whether and, if so, how medicines should be subsidised in Australia. It considers submissions in this context. A PBAC decision not to recommend listing or not to recommend changing a listing does not represent a final PBAC view about the merits of the medicine. A company can resubmit to the PBAC or seek independent review of the PBAC decision.
13. Sponsor’s Comment
Pfizer Australia acknowledges that the reimbursement of new treatments for neuropathic pain is difficult as there is inherent uncertainty in the available evidence and agrees that there is a high unmet need. Therefore, the company will continue to work with the PBAC to make Lyrica available for this population of patients
