  |
|
|
||||
|
||||||
| Journal of the New Zealand Medical Association,
12-September-2003, Vol 116 No 1181
|
||||||
|
Drotrecogin alfa (recombinant human activated protein C) in
severe sepsis – a New Zealand viewpoint
Janet Liang, Stephen Streat, John Torrance, Jamie Sleigh,
Ross Freebairn and Mace Ramsay
Sepsis is a serious intensive care problem
worldwide.1 The incidence of sepsis in
intensive care units is reported to be
increasing2 and New Zealand experience is in
keeping with this.3 Most recent intensive care
unit (ICU) sepsis studies report mortality of around 25–50% but many
factors, including the extent of comorbidity, the nature and site of infection,
adequacy of surgical and antimicrobial therapy, and the severity of the acute
illness, influence the outcome.1 Sepsis was the
reason for ICU admission in 349 of all 1404 ICU deaths (25%) in New Zealand ICUs
in one year.4
The accepted general principles of the treatment of severe
sepsis are to support oxygen transport,5 to
identify and if possible remove the septic source,6,7
and to provide appropriate antimicrobial
therapy.3,8–11 A firm consensus on the
place of adjunctive therapies is not yet established despite considerable
research and recent promising reports, including the PROWESS study of
drotrecogin alfa (recombinant human activated protein
C).12 This agent
(Xigris™, Eli Lilly) is the first
adjunctive therapy for sepsis to be licensed in many countries including the US,
UK, Europe, Australia and, most recently, New Zealand. The agent is expensive
and intensivists have expressed concerns about the benefits, risks, and
financial implications of its use. These concerns led the New Zealand Region of
the Australian and New Zealand Intensive Care Society to request an advisory
statement on the agent from a working party of New Zealand intensivists –
the authors of this paper.
MethodsWe reviewed the original
(PROWESS) study of drotrecogin alfa recombinant in
sepsis,12 the submission made by the sponsoring
company to the (US) FDA,13 more recent
discussions in the New England Journal of
Medicine,14–16 a Canadian economic
evaluation of the use of the agent,17
Australian pharmaceutical benefits scheme positive
recommendations,18 guidelines for the use of
drotrecogin alfa in sepsis produced by the Eli Lilly Australian Advisory Board
(personal communication, M Fisher, 2002), positive (personal communications, M
O’Leary, J Reeves, Y Shehabi, G Dobb, G Skowronski, R Herkes, J Lipman, D
Stephens, JW Mulder, D Cook, 2002), negative (personal communications, M Fisher,
JF Cade, P Harrigan, J Santamaria, 200) and still pending (personal
communications, B Richards, M Parr, D Milliss, 2002) Australian
hospital-pharmacy-committee decisions on availability and New Zealand pricing
and payment arrangements. We then formulated consensus guidelines for the use of
this agent in New Zealand.
ResultsReview
of relevant information The PROWESS study (a multicentre controlled trial
in 1690 randomised adult patients with severe sepsis, published in March 2001)
showed a 6.1% absolute reduction (30.8% to 24.7%, p = 0.005) in all-cause 28-day
mortality from a 96-hour infusion of drotrecogin alfa at 24 ug/kg body
weight/hour, despite a possibly increased risk of serious bleeding (3.5% versus
2.0%, p = 0.06) in the drotrecogin alfa
group.12 The company that sponsored the study
(Eli Lilly) sought product registration in the United States in September 2001
and provided the FDA with extensive
documentation13 but only one phase III
randomised controlled trial. Despite approving the agent for use for
‘insert indication’, the FDA Anti-infective Drug Advisory Committee
was split 10 to 10 as to whether the agent is safe and
efficacious.16 The ‘key matters of
concern’ for the FDA were changes made during the trial, the use of APACHE
II scores,19 and the risk of serious bleeding.
The decision to approve the agent despite these concerns has recently been
discussed and defended by a senior FDA
member.15 Concern over inconsistency of the
efficacy of the agent throughout the trial (with possible implications that the
mid-trial protocol amendment or changes in the formulation of the agent were
responsible) has led to a recently expressed
view16 that the data at present do not provide
sufficient evidence for the use of the agent to become ‘the standard of
care’. In subsequent correspondence14 the
PROWESS authors discuss these concerns (but do not refute them directly) and
suggest that ‘clinicians can already incorporate level I evidence from
PROWESS into their practice to obtain life-saving benefit for their
patients’. A recent Canadian economic analysis suggested that the cost per
life-year gained by treatment with the agent was US$27 936 if all eligible ICU
patients are treated, and US$24 484 if only patients with APACHE II scores of 25
or more are treated.17 Furthermore, if patients
with such high APACHE II scores are treated, the cost per life-year gained was
related to the age of the patient (US$16 309 aged over 40, US$28 100 aged 80 or
more). The Australian PBAC recommendations18
were that the agent be ‘Recommended for listing for ‘Adult patients
with severe sepsis who have a high risk of death as determined by acute organ
dysfunction in at least two organs or modified APACHE II score of at least
25’ on the basis of acceptable cost-effectiveness’, and be
‘restricted to patients with two or more failed organs to prevent use in
less severely ill patients where the risks may outweigh the benefits’. The
Eli Lilly Australian Clinical Advisory Board recommendations were in keeping
with the PBAC recommendations but included a recommendation that assessment of
progress be made after four hours’ full resuscitation in an intensive care
unit, including surgical therapy and antibiotics and that ‘if objective
improvement in organ function occurs’ that administration ‘be
delayed’. This Board also recommended that, for the purposes of defining
respiratory ‘organ dysfunction’, this should be due to ‘lung
injury/ARDS secondary to sepsis’ and that any patient given the agent
would be expected to be receiving ventilatory support. Several tertiary
Australian hospitals have not approved the use of the agent because of the
‘extreme financial implications’. The New Zealand price of the agent
(to a hospital pharmacy) is currently $1909 plus GST per 20 mg vial (personal
communication, Eli Lilly, 2002), which would result in a cost of $17 181
(including GST) for the treatment (total 160 mg) of a 70 kg patient. Hospitals
wishing to use the agent will have to find this cost from within existing
budgets.
Suggested guidelines for
the use of drotrecogin alfa in New Zealand If the agent is made available
for use, we recommend that specialists prescribing it be required to contribute
clinical data to a national register of patients, and we recommend the following
guidelines for its use:
DiscussionIt is likely that there is an
overall beneficial therapeutic effect (net reduction in 28-day mortality) from
the use of drotrecogin alfa recombinant in selected intensive care patients with
severe sepsis but this is debated.16 The size
of this effect (and thereby the number of patients needed to treat) is debatable
in the New Zealand clinical context. The PROWESS data suggested an NNT of 16.4
(95% confidence limits 9.6–52.6)12 to
result in one additional 28-day survivor. The increased risk of bleeding in the
study was small in patients selected not to have high risk of
bleeding.
Of particular concern in the New Zealand (and Australian)
context is the problem of young patients with severe meningococcal disease, many
of whom have at least moderate coagulopathy and would thereby have been excluded
from the PROWESS study.12 Although these
patients may benefit from the agent, they are almost certainly at higher than
usual risk of bleeding and a cautious approach to treatment is advised. To date,
there have been only five reported cases (age 18–41, median 22) where
drotrecogin alfa was used in meningococcal purpura
fulminans.20,21 Profound thrombocytopaenia
before treatment with drotrecogin alfa was present in two patients and these
were given platelet transfusions ‘to maintain platelet count above 30 x
109 per litre’. No adverse bleeding
events were reported. Children under 18 were excluded from the PROWESS study and
there are no published randomised controlled trials in children with severe
sepsis, although one is underway.22 The median
time till death in a small series of children dying of meningococcal disease in
Auckland was four hours and thrombocytopaenic cerebral haemorrhage was a
significant cause of late deaths (personal communication, J Beca, 2003). In view
of the strong association of profound thrombocytopaenia with mortality in young
children with meningococcal disease,23 we
recommend a cautious approach to the use of drotrecogin alfa in such
patients.
The issue of very high cost, moderately effective treatment
is not just one for intensive care. The price of drotrecogin alfa is large and
the resultant cost per life-year gained is of similar order to that of a small
number of other treatments (eg, imatinib
(Glivec®) for liver transplantation,
iloprost for pulmonary hypertension) that in New Zealand are subject to
rationing. Drotrecogin alfa was given provisional consent in New Zealand on 19
September 2002.24
Although New Zealand ICU
practice25,26 probably differs from Canadian
practice17,27 in casemix and approach, the
relative cost-benefit implications of age and illness severity will remain
relevant. Possible suggested strategies that might increase the cost
effectiveness of this agent include restricting it to patients who are not
‘clearly improving’ after six hours of ‘full intensive care
therapies’ in an ICU and restricting its use to younger patients with high
severity of illness17 (eg, APACHE II over 25)
who do not have limiting non-septic comorbidity. We support these strategies in
our recommendations for use.
The responsibility for providing access to this agent is
that of individual hospitals and this decision is expected to fall on hospital
pharmacy committees. We recommend that individual area health-board pharmacy and
therapeutics committees review the agent now and decide whether or not they will
support its purchase and use.
Finally, we note that timing and appropriateness of
surgical6 and
antibiotic8–11 therapy and resuscitation
of oxygen delivery5 are powerful determinants
of outcome in severe sepsis and suggest that all hospitals would be well advised
to formally establish systems that ensure these factors of treatment are
provided.
Author information:
Janet Liang, Clinical Director, Intensive Care Unit, North Shore Hospital,
Takapuna; Stephen Streat, Intensivist, Department of Critical Care Medicine,
Auckland Hospital, Auckland; John Torrance, Clinical Director; Jamie Sleigh,
Intensivist, Intensive Care Unit, Waikato Hospital; Ross Freebairn, Clinical
Director, Intensive Care Unit, Hawkes Bay Hospital; Mace Ramsay, Clinical
Director, Intensive Care Unit, Dunedin Hospital, Dunedin
Correspondence: Dr
Janet Liang, Intensive Care Unit, North Shore Hospital, Private Bag 93-503,
Takapuna. Fax: (09) 489 0522; email: jathga@internet.co.nz
References:
Comment:
Drotrecogin alpha (activated): a magic bullet or budget blowout?For the first time since antibiotics
were introduced for the treatment of infection, an adjunctive therapy has been
shown to improve survival. In the presence of sepsis, drotrecogin alpha
(recombinant human activated protein C) modulates the systemic inflammatory,
procoagulant, and fibrinolytic host responses to
infection.1 The PROWESS multicentre study in
which drotrecogin alpha (activated) was randomised to 1690 patients showed an
absolute all-cause 28-day mortality reduction of 6.1%
overall.1 However, the benefits were most
marked in more severe sepsis (Acute Physiology and Chronic Health Evaluation
(APACHE) II score >25).2
This treatment is not cheap. The cost of a 96-hour treatment
is NZ$17 181.3 If one uses the PROWESS entry
data, for each life saved 16 patients need to be treated. Drotrecogin alpha
(activated) costs US$160 000 (NZ$278 000) per life saved, but as little as US$27
400 (NZ$48 000) per quality-adjusted life-year when limited to patients with an
APACHE II score ≥25.2
How does this stack up against other high-cost therapies?
The cost effectiveness of drotrecogin alpha (activated) is comparable to most of
the interventions in Figure 1 and better than that of airbags, implantable
defibrillators, lung transplantation, and cardiopulmonary
resuscitation.
Figure 1. Comparison of drotrecogin alpha with other
widely used interventions, NZ$ per quality-adjusted life-years (adapted from
Figure 6, reference 2) (click here to
view)
In selected patients this therapy is clearly cost effective
and beneficial. The problem is that New Zealand’s public hospital system
is fragmented into 21 district health boards and (unlike our Australian, UK, and
US counterparts) concentrates on cost cutting as opposed to efficiency gains. In
an environment where the funding for drotrecogin alpha (activated) is dependent
on hospital pharmaceutical budgets it seems doubtful that all New Zealanders who
may benefit from this treatment will get it. Inequity of access is at odds with
Right 4 (3, 4) of the Health and Disability Code of
Rights:4
In the US, drotrecogin alpha (activated)
has been granted new-technology status from the Centers for Medicare and
Medicaid Services (CMS). This allows hospitals to receive additional
reimbursement for treatment of Medicare
patients.5 Similarly, New Zealand should fund
drotrecogin alpha (activated) nationally, with the conditions of that funding
based upon agreed guidelines, and at the same time develop a central database.
This would improve equity of access and allow audit of the impact of the use of
drotrecogin alpha (activated) on both patients and budgets.
Geoffrey M Shaw
Clinical Senior Lecturer, Christchurch School of Medicine and Health Sciences, University of Otago Intensivist, Department of Intensive Care, Christchurch Hospital, Christchurch References:
|
|
||||
| | Current
issue | Search journal |
Archived issues | Classifieds
| Hotline (free ads) Subscribe | Contribute | Advertise | Contact Us | Copyright | Other Journals |
|