Methylnaltrexone bromide, injection, 12 mg in 0.6 mL (base), Relistor®, March 2009
Public summary document for Methylnaltrexone bromide, injection, 12 mg in 0.6 mL (base), Relistor®, March 2009
Page last updated: 17 July 2009
Public Summary Document
Product: Methylnaltrexone bromide, injection, 12
mg in 0.6 mL (base), Relistor®
Sponsor: Wyeth Australia Pty Ltd
Date of PBAC Consideration: March 2009
1. Purpose of Application
The submission sought an authority required listing in the
Palliative Care Section for initial and continuing treatment of
opioid-induced constipation in patients who have failed to respond
to, or are unable to tolerate laxative therapies.
2. Background
This drug had not previously been considered by the PBAC.
3. Registration Status
Methylnaltrexone (MNTX) was TGA registered on 17 November 2008 for
the treatment of opioid-induced constipation in patients with
advanced illness who are receiving palliative care when response to
laxative therapy has not been sufficient.
4. Listing Requested and PBAC’s View
Authority Required
Opioid-induced constipation. Patients must have failed to respond
to, or are unable to tolerate, laxative therapies.
Authority Required
Opioid-induced constipation. Patients must have demonstrated a
response to previous doses of methylnaltrexone.
The sponsor, in its pre-Sub-Committee Response, accepted the
following amendments to the restriction:
Authority Required
Initial supply for palliative care patients with opioid-induced
constipation who have failed to respond to, or are unable to
tolerate, laxative therapies.
Continuing supply for palliative care patients with opioid-induced
constipation who have demonstrated a response to methylnaltrexone,
and where consultation with a palliative care specialist or service
has occurred.
NOTE:
No applications for increased repeats will be authorised.
Authority Required
Continuing supply for palliative care patients with opioid-induced
constipation who have demonstrated a response to
methylnaltrexone.
NOTE:
No applications for increased repeats will be authorised.
For PBAC’s view, see Recommendation and
Reasons.
5. Clinical Place for the Proposed Therapy
Opioids form the mainstay of pain management for patients receiving
palliative care. However, opioid therapy is often associated with a
range of adverse effects, including opioid-induced constipation
(OIC). In patients with cancer, constipation is a common
complication and a cause of significant distress. In addition to
receiving opioid therapy, palliative care patients are also
immobile, may have a diet lacking in fibre and a reduced fluid
intake, all of which can contribute to constipation.
Currently the mainstay of OIC management is the use of oral
laxatives for symptomatic treatment. Laxatives are usually
prescribed upon initiation of opioid therapy, however a proportion
of patients become unresponsive to, or are unable to tolerate, oral
laxative treatment. Many laxative agents require significant fluid
intake, which can be difficult to tolerate for a patient with
advanced stage disease. When patients no longer respond to oral
laxatives, topical interventions include enemas and suppositories,
and manual disimpaction may be required in some patients.
MNTX would provide an alternative treatment for opioid-induced
constipation in palliative care patients where the patient cannot
tolerate or is not responding to other available treatments.
6. Comparator
The submission nominated placebo as the main comparator. The PBAC
considered this was appropriate.
7. Clinical Trials
The submission was based on two direct randomised controlled trials, comparing methylnaltrexone and placebo:
- MNTX 301 was a multi centre, single dose, randomised, controlled study (either 0.15 mg/kg or 0.30 mg/kg methylnaltrexone vs placebo), followed by a 28 day variable dose, open label period (dose varied at clinician’s discretion). MNTX 301 EXT was a 3 month, variable dose, open label extension of MNTX 301 (dose varied at clinician’s discretion).
- MNTX 302 was a 14 day, randomised, controlled study (0.15mg/kg methylnaltrexone vs placebo). MNTX 302 EXT was a 3 month, variable dose, open label extension of MNTX 302 (dose varied at clinician’s discretion).
The submission is primarily based on MNTX 302.
The trial published at the time of submission is listed below:
| Trial ID | Protocol title/ Publication title | Publication citation |
| MNTX 302 Thomas et al. | Methylnaltrexone for Opioid-Induced Constipation in Advanced Illness | New England Journal of Medicine, 2008: 358: 2332-43 |
8. Results of Trials
The MNTX 301 trial compared methylnaltrexone 0.15 mg/kg and 0.30
mg/kg to placebo for laxation within 4 hours and 24 hours of a
single subcutaneous administration. The MNTX 302 trial compared
methylnaltrexone 0.15 mg/kg to placebo for laxation within 4 hours
and 24 hours of each of 7 administrations in a 14 day period. All
patients continued their usual laxative therapy during the trials.
Patients requiring rescue therapies (invasive interventions such as
suppositories, enemas or manual evacuation) to achieve laxation
were treated as non-responders.
The trial results of proportion of responders for rescue-free
laxation within 4 hours of first dose in trials MNTX 302 and MNTX
301 showed that there were statistically significantly more
responders in the methylnaltrexone arms compared to placebo. There
was no evidence of an increase in efficacy with the higher dose of
methylnaltrexone (0.30 mg/kg) compared to the lower dose (0.15
mg/kg).
The proportion of responders for rescue-free laxation within 4
hours of dose and 24 hours of dose over 7 doses in trial MNTX 302
was consistently greater than in those patients receiving placebo.
There was no apparent attenuation of efficacy over 14 days of
treatment.
There was no evidence of attenuation of efficacy for
methylnaltrexone over the MNTX 302 EXT study period of 3 months,
with surviving RCT treatment arm patients maintaining their
proportion of responses to treatment and surviving RCT placebo arm
patients attaining similar response proportions to those from the
treatment arms.
Quality of life was not assessed in either MNTX 302 or MNTX 301 or
associated extension studies.
During the randomised placebo controlled trials and extension
studies (including the 30 day follow up periods) 56% of patients
died in study MNTX 301 and 45% of patients died in study MNTX 302.
One death due to dehydration secondary to severe diarrhoea was
attributed to methylnaltrexone which occurred when this patient was
given the higher dose of 0.30 mg/kg (MNTX 301 open-label study).
Dehydration was reported less frequently in the methylnaltrexone
group compared to the placebo group in the double-blind study of
trial MNTX 302 and there was no statistically significant
difference between treatment and placebo groups in reports of
diarrhoea (p = 0.479).
In trials MNTX 301 and MNTX 302 treatment related adverse events
were reported more frequently in the methylnaltrexone groups
compared to placebo. However, more severe adverse events were
reported in the placebo groups (MNTX 301 = 9.6%, MNTX 302 = 28.2%)
compared to the methylnaltrexone groups at a dose of 0.15 mg/kg
(MNTX 301 = 4.3%, MNTX 302 = 23.8%). The most frequently reported
severe adverse event was neoplasm progression. Adverse events and
severe adverse events were more frequent in the open-label
extension studies (MNTX 301 = 96.7%, MNTX 302 = 100%) compared to
the double-blind studies.
The adverse events reported more frequently in the treatment groups
compared to placebo were abdominal pain, flatulence, nausea,
vomiting, diarrhoea and dizziness.
9. Clinical Claim
The submission claimed that methylnaltrexone is superior to placebo
in terms of comparative effectiveness for the treatment of
opioid-induced constipation in palliative care patients and that
methylnaltrexone is generally well tolerated with the occurrence of
treatment emergent adverse events similar to placebo.
The PBAC accepted this claim, however noted that patients receiving
methylnaltrexone reported more adverse events than those receiving
placebo due to an increased incidence of gastrointestinal pain and
discomfort.
10. Economic Analysis
The submission presented a Markov analysis based on a stepped
economic evaluation with a time horizon of 4 months (extrapolated
from the 2 week MNTX 302 RCT data) and a cycle length of one day
branching to five health states; optimal (laxation within 4 hours),
moderate (laxation between 4 and 24 hours), and non-responders (no
laxation), dropouts and deaths.
The economic evaluation resulted in a baseline ICER in the range of
$15,000 - $45,000 per QALY. The results of the sensitivity analyses
indicated that the model was most sensitive to assumptions
regarding frequency of dosing, the utility of optimal responders
(no constipation), and optimal/moderate responder rates
respectively.
The submission derived health related quality of life utility
weights for opioid induced constipation from Schmier et al. (2002)
and the MNTX 302 trial and extension study data.
For PBAC’s comments, see Recommendation and
Reasons.
11. Estimated PBS Usage and Financial Implications
The financial cost per year to the PBS was estimated to be less
than $10 million in Year 5. This was considered to be an
underestimate.
12. Recommendation and Reasons
The PBAC considered that the proposed restriction for
methylnaltrexone did not define those patients at highest risk and
that any future submission should consider restricting use to a
subgroup of patients with high clinical need, such as end stage
palliative care patients. The PBAC also considered that treatment
should be limited to a maximum period of 4 months, given that the
evidence supporting the efficacy and safety of methylnaltrexone is
limited to a time horizon of 4 months. Pre-filled syringes should
also be available in preference to the vials as this would
facilitate ease of use for carers managing patients at home and
potentially reduce wastage.
The PBAC noted comments from consumers and acknowledged that
opioid-induced constipation (OIC) is a significant problem. The
PBAC agreed that methylnaltrexone is an effective medication for
OIC. The PBAC considered that the clinical claim that
methylnaltrexone is superior to placebo in terms of comparative
effectiveness for the treatment of opioid-induced constipation in
palliative care patients was reasonable. However, it was noted that
patients receiving methylnaltrexone reported more adverse events
than those receiving placebo due to an increased incidence of
gastrointestinal pain and discomfort.
The PBAC noted that the submission presented a Markov analysis
based on a stepped economic evaluation with a time horizon of 4
months (extrapolated from the 2 week MNTX 302 randomised control
trial data). This was considered reasonable as the responses seem
to be maintained.
However, the PBAC identified 3 main areas of concern in the
economic model.
Firstly, the economic model aligns constipation states with
response categories to methylnaltrexone from Study MNTX 302 (2 week
data) and uses 3 patient groupings (severe constipation equates to
laxation in 24 hours; moderate constipation equates to laxation in
4-24 hours; no constipation equates to laxation within 4 hours )
and assumes that all patients start in the severe constipation
group. However, there is no basis provided to support this
alignment of different concepts of health state levels and response
categories – it is not likely to be the case that the
physical act of laxation relieves all the feelings and symptoms of
constipation such as bloating and abdominal discomfort. Further,
the PBAC noted that all patients continued their usual laxative
therapy during the trials (MNTX 301 and 302). Patients requiring
rescue therapies (invasive interventions such as suppositories,
enemas or manual evacuation) to achieve laxation were treated as
non-responders.
Secondly, the transformation of the mean utility scores for
constipation health states derived from Schmier et al. (2002) to
weighted utilities for constipation treatment response is not
adequately justified in the submission. Again, there is no basis
provided to support this alignment of different concepts of health
state levels and utilities. The PBAC considered that, overall it
was not reasonable to assume that the utility of an optimal
treatment responder is the same as the utility of someone without
constipation, nor that the utility of a moderate responder is the
same as the utility of a person with moderate constipation. The
PBAC noted that there is considerable uncertainty associated with
these assumptions and the model is sensitive to changes in these
utilities.
In this regard, the PBAC further noted that the Schmier study
estimated preference weights, which are only interpretable relative
to each other, and not in any other context, and not as acceptable
QALY weights. It appears that preference weights have been derived
from the utility functions estimated based on the choice experiment
conducted in the Schmier study, and these have been fitted to a 0-1
scale, but insufficient information about the methods is provided
in the original paper and in the submission to explain the methods.
If the estimates are the coefficients from the regressions, it
should be noted that utility functions estimated from choice
experiments are subject to scale effects, and it is not appropriate
to interpret this as a cardinal utility function on a 0-1 scale.
The tradeoffs considered in this study can only be interpreted
relative to each other, eg how many more days of pain to avoid
constipation. The PBAC considered that for a preference index to be
interpreted as a QALY weight, there needs to be some trade-off of
survival and quality of life (QOL) presented in the study from
which the estimates are derived – which is not what has been
undertaken by the submission because all scenarios are of 2 weeks
duration. There is no evidence that the scale is anchored at 0 =
death and 1 = full health, and, in the context of the listing
sought, the utility increments across the three health states seem
implausibly large when considered as the amount of remaining life
being traded off for each increment. The PBAC noted that no
information is provided on the actual model (type of model, the
model specification, what beta parameters were included, any
parameters of fit, etc) used to derive these values and that there
is no information on the design of the conjoint analysis. In
addition the structure of the modelled economic evaluation does not
fully capture the two clinically relevant outcomes: laxation within
4 hours (primary outcome in the trials) and more than 24 hours
after therapy, both of which are relevant to patients.
The PBAC also considered that there is further uncertainty in the
utility estimates due to possible differences in populations (i.e.
trials, Schmier and Australian populations) and uncertainty
regarding how the adaptive conjoint analysis selected the scenarios
to give to participants. The PBAC noted that, in the
methylnaltrexone trials the majority of patients were classified as
being WHO performance status of 3 or 4, meaning they were
significantly less mobile than the population in the Schmeir study
that attended clinics to complete the interviews for the conjoint
analysis. In the Schmeir study, only 20% of patients had cancer and
this had not necessarily reached a terminal phase. The PBAC
considered that the supplementary analysis provided by the sponsor
in its Pre-Sub-Committee response did not satisfy the notion that
these are now similar populations that would elicit similar
responses in valuing their relative health states.
Thirdly, the results of the sensitivity analyses indicate that the
model is most sensitive to the frequency of methylnaltrexone
dosing, the utility of optimal responders (no constipation), and
optimal/moderate responder rates. The PBAC noted that if doses are
given on alternate days rather than at the longer intervals (3.8 to
4.3 days) reported in trial MNTX 302 EXT, this produces an ICER in
the range of $45,000 - $75,000 /QALY compared to the baseline which
was in the range of $15,000 - $45,000. Using the utility of mild
constipation in Schmier of 0.64 for the optimal responder utility
produces an ICER in the range of $45,000-$75,000/QALY. Varying
response rates for optimal and moderate responders by 95%
confidence intervals produced an ICER for which the lowest value
was in the range of $15,000 - $45,000 and the highest value was in
the range of $45,000 - $75,000/QALY for optimal responders and
moderate responders. A multi-variate analysis, assuming alternate
day administration and an optimal responder utility of 0.64
produces an ICER in the range of $75,000 - $105,000 /QALY. The PBAC
considered that given the sensitivity of the model to changes in
these utilities, there is considerable uncertainty associated with
these estimates.
The PBAC considered that there was significant uncertainty about
the size of the eligible population due to variation in the
reported incidence of constipation and failure with existing
PBS-listed laxative treatments defined as rescue “invasive
procedures” (ie suppositories and enemas). The PBAC noted
that there was also considerable uncertainty around the uptake of
the drug in the eligible population and that there was considerable
risk that methylnaltrexone would be used beyond the restricted
population (as first line treatment for constipation in the
palliative care population and in patients with chronic disabling
pain requiring long term opioid therapy).
The PBAC noted that the utilisation estimates were based on the
availability of a single strength 12 mg vial which could result in
considerable wastage in patients weighing less than 62 kg and that
the availability of 8 mg pre-filled syringes could result in cost
savings from less wastage.
Therefore, the PBAC rejected the application on the basis of high
and uncertain cost-effectiveness.
13. 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.
14. Sponsor’s Comment
Wyeth acknowledges the PBAC comments and will continue to work with
the PBAC to ensure a successful listing for the benefit of
palliative care patients.
