Microsoft word - ulipristal _ellaone_ for post-coital contraception - netag appraisal report, sept09.doc

Ulipristal (ellaOne®)
for post-coital
contraception
• Ulipristal is a first-in-class progesterone receptor modulator licensed for post coital (‘emergency’) contraception up to 120 hours (5 days) following unprotected intercourse. • Ulipristal has demonstrated efficacy in non-comparative and randomised comparative studies versus levonorgestrel. The results indicate that it is at least as effective as levonorgestrel. • The adverse effect profile for ulipristal is similar to that seen with levonorgestrel, with mild gastrointestinal symptoms and disruption to normal menstrual patterns the most frequently occurring and common effects. The effects of inadvertent ulipristal exposure in utero are not known. • Ulipristal is about three times the cost of levonorgestrel at £16.95 per dose versus £5.37 per dose. • Levonorgestrel is widely provided within NHS North East. It is licensed for use up to 72 hours (3 days) following unprotected intercourse. It is available from several sources, many without the need for prescription. Ulipristal will initially be available on North East Treatment Advisory Group. September 2009 Background and introduction

Emergency contraception is the lay term applied to any contraceptive
intervention that is made post-coitus (more accurately referred to as post-coital
contraception [PCC]). There are many different reasons why PCC may be
required. Current treatment options consist of either:
• Levonorgestrel 1.5 mg (Levonelle®) • or, Insertion of an intra-uterine device (IUD) Levonorgestrel is considered the first-line treatment option compared with an IUD because of its ease of access (it can even be purchased without a prescription), ease of administration requiring no third party assistance and being non-invasive, an acceptably high level of efficacy, and relatively low cost. Levonorgestrel is taken as a single dose as soon as possible, and no later than 72 hours, post-coitus. If PCC is required beyond the 72 hour window afforded by levonorgestrel then an IUD can be inserted up to five days post-coitus. Levonorgestrel is estimated to prevent about 85% of conceptions as it is used in practice. However it is also widely considered to be more effective in the first 24 hours (95%) and less effective in the last 24 hours of that period (58%) so that expediency of consumption is advised. An IUD is 100% effective at whatever point in the five days post-coital period it is inserted. In practice, levonorgestrel is often used ‘off-license’ for PCC up to 120 hours (5 days), although it is not known what level of efficacy it provides beyond 72 hours. 1 Normal sperm can survive in the female reproductive tract and retain the ability to fertilize an egg for at least three, and up to five, days. An oocyte can be successfully fertilised for only approximately 12 to 24 hours after it has been released. Although not specifically stated, it is assumed that for this reason several published studies of PCC in women are conducted up to a maximum of 120 hours post-coitus. 2,3 Ulipristal (ellaOne®) is a first-in-class progesterone receptor modulator licensed for contraception up to 120 hours (five days) post-coitus. It is taken as a single 30 mg dose, containing micronised drug particles. 4 Its mode of action has not been fully elicited but its principal mode of action is believed to be via progesterone receptor antagonism thereby preventing the pro-pregnancy effects of progesterone. Specifically, it results in delayed ovulation. Ulipristal has also demonstrated some progesterone agonistic effects in vitro. 5-7 Ulipristal has been investigated for a number of other indications including prospective contraception and uterine fibroids/leiomyoma. 5-7 The purpose of this report is to appraise the evidence for the safety and efficacy of ulipristal when used for contraceptive purposes, advise on the potential financial implications and inform a recommendation on the use of ulipristal to be made by the North East Treatment Advisory Group. North East Treatment Advisory Group. September 2009 Clinical evidence

Two key phase II studies 3,8 and two phase III studies 3,9 of the use of ulipristal as
a method of PCC have been completed. Different formulations of ulipristal were
used in the studies with initial research utilising a 50 mg tablet, whereas the
phase III studies used a 30 mg tablet containing micronised drug particles. The
two formulations are considered to be bioequivalent. 3 One phase II and one
phase III study were comparative studies versus levonorgestrel. The data from
these two studies has been combined in an as yet unpublished meta-analysis.
In the only published evidence, a phase II study, randomised 1,672 women
requiring PCC within 72 hours to treatment with ulipristal 50 mg (n = 832) or
levonorgestrel 1.5 mg taken as two 750 microgram doses approximately 12
hours apart (n = 840). 8 Patients had a mean age of 24 years (minimum 18),
were predominantly white (73%), and well educated (86% with college or higher
education). Efficacy results are based on 792 and 786 patients respectively with
omissions due only to loss to follow-up. The trial was designed to demonstrate
the non-inferiority of ulipristal compared with levonorgestrel. Twenty-six
pregnancies occurred, 12 in patients treated with ulipristal and 14 with
levonorgestrel, corresponding to rates of 1.5% and 1.8% respectively (p > 0.05).
Based on pre-treatment blood test results and ultra-sound scans, four
pregnancies in the ulipristal group and one in the levonorgestrel group were
subsequently found to have occurred prior to study entry. Exclusion of these
patients plus patients who had used additional PCC within the same cycle
resulted in 7 (0.9%) and 13 (1.7%) pregnancies respectively. Ulipristal was
calculated to have prevented 85% of expected pregnancies compared with 69%
in levonorgestrel patients (p > 0.05).
The other phase II study of ulipristal for PCC is not published. The following
results have been obtained from the European Medicines Agency public
assessment report (EPAR) where the study is referred to as HRA2914-508. 3
Four hundred patients (mean age 24 years, 70% white) were randomised to PCC
treatment with either ulipristal 50 mg or ulipristal (micronised) 10 mg, taken within
72 hours of coitus. The micronised formulation was found to be non-significantly
inferior to the standard preparation with an estimated 52% of pregnancies
prevented compared with 76%.
The phase III studies are also unpublished and again results are obtained from
the EPAR and additionally unpublished data from the license holder. The first
study is referred to as HRA2914-509 and was an open-label non-comparative
study of ulipristal 30 mg (micronised) taken within 48 to 120 hours post-coitus. 3
1,623 patients were enrolled, 1,533 were treated and included in the intention-to-
treat population. Within the intention-to-treat population there were 29
pregnancies, yielding a rate of 2.17% compared with an expected rate of 5.64%.
North East Treatment Advisory Group. September 2009 The second phase III study is referred to in the EPAR as HRA2914-513. 3 Data is
currently available only from conference abstracts and unpublished commercial
literature. 9,10 In this study patients were randomised to ulipristal 30 mg (n = 939)
or levonorgestrel 1.5 mg (n = 954) within 120 hours of coitus. In the whole patient
population there were 15 pregnancies in the ulipristal group (1.6%) and 25 in the
levonorgestrel group (2.6%). The difference was not significant (p > 0.05). In the
subset of patients treated up to 72 hours (n = 843 and 851 respectively) there
were 15 pregnancies in the ulipristal group (1.8%) and 22 in the levonorgestrel
group (2.6%). The difference was not significant (p > 0.05). Non-inferiority of
ulipristal compared with levonorgestrel was adequately demonstrated.
A meta-analysis of the phase II and phase III comparative studies, each involving
different formulations of ulipristal and a different regimen for levonorgestrel, was
also presented at the same conference in September 2009. 10,11 The data is
currently available in abstract and cannot be fully appraised. The evidence shows
that over the 120 hour period ulipristal is significantly more effective than
levonorgestrel at producing a reduced pregnancy rate.
Safety

Data relating to the adverse effects of ulipristal are not readily available as many
of the studies have not yet been published. In the published phase II comparative
study the incidence of adverse effects appears to be similar in both groups. The
most common observed adverse effects are headache, nausea, vomiting, and
mild abdominal pain. 8
In the phase III comparative study the most common adverse effects with
ulipristal and levonorgestrel were, respectively; headache (both 19%),
dysmenorrhea (13% and 14%), nausea (13% and 11%), abdominal pain (5% and
7%), dizziness (both 5%), fatigue (6% and 5%), and upper abdominal pain (3%
and 4%). 10
Effects resulting in disruption to the normal menstrual cycle appear common. It
has been observed that ulipristal results in an increase to the duration of the
average menstrual cycle of two days. This would be expected with its primary
mode of action believed to be due to ovulatory delay. 3 Additionally, the product
data sheet provides the following statement: 4
“The majority of women (80.8%) in the phase III study had their next menstrual period at the expected time or within ± 7 days, while 6.1% experienced menses more than 7 days earlier than expected and 19.2% had a delay of more than 7 days beyond the anticipated onset of menses. This delay was greater than 20 days in 5.1% of the women, and 0.5 % of women experienced a delay of more than 60 days beyond the anticipated onset of menses. Menstrual volume was reported by 79.0% of women as normal, 16.0% as heavy and 5.0% as spotting. A minority (8.7%) of women reported intermenstrual bleeding lasting an North East Treatment Advisory Group. September 2009 average of 2.4 days. In a majority of cases (91.8%), this bleeding was reported as spotting. Among the women who received Ellaone in the phase III study, only 5 (0.3%) reported heavy intermenstrual bleeding.”
Little is known about the long term effects of ulipristal but in clinical use patients
would be expected to be exposed to only small cumulative life time doses with
long latent periods between doses. A three month study exposed healthy female
patients to low daily doses (2.5, 5 or 10 mg). This found no significant increase in
adverse effects compared with placebo other than an increased incidence of
acne for patients taking 5 or 10 mg of ulipristal. Other common effects included
headache (~40%), dysmenorrhea (~20%), breast pain (~10%), and pelvic pain
(~10%). There were no blood pressure changes or weight gain, and no changes
observed with liver function tests, electrolyte screens, creatinine levels, blood
sugar levels, lipid screens, cortisol, prolactin, or testosterone levels. 12
One important aspect that has not been adequately addressed is the outcomes
of pregnancies that are inadvertently exposed to ulipristal. Most instances of
inadvertent exposure would likely occur in the first trimester, and indeed within
the first few weeks of the first trimester, a period during which embryonic
development is particularly sensitive. The EPAR states that of four known births
following in utero ulipristal exposure at various statim doses all were normal.
Summary of evidence

There are concerns of significant methodological flaws in the calculation of
expected pregnancy rates as this relies on the theory of a six-day fertile window
per cycle identified with reference to the point of ovulation. However the time of
ovulation is not known and is therefore assumed to be fourteen days prior to
menstrual bleeding for women with consistent cycle duration. All studies recruited
only women who reported consistent cycles. This problem is common to all
studies of pharmacological contraception and there is little that can be done to
overcome it in practice. 3 This places even greater value on data from the
randomised comparative studies.
The evidence base for ulipristal demonstrates that it confers an acceptably high
rate of efficacy which appears to be at least as effective as levonorgestrel.
However only one study has been published and evidence from the remaining
studies is taken from a secondary source and unpublished sources such as
conference abstracts and commercial literature. The published comparative
study made use of a formulation of ulipristal that will not be available, and an
obsolete regimen of levonorgestrel. Both are considered to be equivalent to the
available preparation and the standard regimen respectively. The safety profile of
ulipristal appears similar to that of levonorgestrel although there is still
uncertainty regarding inadvertent exposure in utero.
North East Treatment Advisory Group. September 2009 Cost analysis

The price of ellaOne® is £16.95 compared with £5.37 for Levonelle®. IUDs
currently available on NHS prescription range in cost from £8 to £27 (median
value £10.53). However, about 15% of Levonelle® primary care prescriptions
within NHS North East are inadvertently prescribed as the over-the-counter
Levonelle-One Step® proprietary brand, at a cost of £13.83. 13,14
In the last financial year (April 2008 to March 2009) nearly 10,000 doses of
levonorgestrel 1.5 mg were dispensed on primary care prescriptions within NHS
North East. 14 Additionally, most areas have in place arrangements for provision
other than via primary care prescriptions (e.g. direct supply from a community
pharmacy or through a patient group directive by a nurse or health visitor). Other
sources of provision at NHS expense include acute hospital trusts (e.g. via an
accident and emergency department), and specialist clinics and centres (e.g.
NHS walk-in centres). A sample of data from six predominantly urban primary
care trusts within NHS North East indicates that levonorgestrel supplied directly
from pharmacies at local NHS expense (i.e. not private purchases; for example
via patient group directives or specially commissioned services) account for a
greater volume than that which is prescribed on primary care prescriptions (i.e.
four times the prescription volume). No data was obtained concerning the volume
of dispensing/supply via other sources.
Unfortunately the total demand within NHS North East for PCC, as measured via
provision of levonorgestrel and IUDs for PCC, cannot be accurately estimated. It
can be stated that demand for PCC is a frequent occurrence especially within
certain patient groups (e.g. young females).
The evidence from clinical studies of ulipristal of up to 120 hours demonstrated
that only about 10% of supplies were made in the period between 72 and 120
hours. 3,8,10 This rate may be subject to patient recall bias and deliberate
manipulation. If this rate is replicated in practice it would mean that 90% of
demand can be satisfied with use of levonorgestrel within its product license. The
remaining 10% could be treated with levonorgestrel ‘off-license’, IUD insertion, or
remain untreated with subsequent sequelae including no pregnancy, termination
or delivery.
North East Treatment Advisory Group. September 2009 Relevant substitution costs depending on treatment choices include: 13,15 a. For each ulipristal prescription instead of a prescription for levonorgestrel b. For each ulipristal prescription instead of a prescription for an IUD an additional £6.00 is incurred (range -£10 to £9) although this will likely be more than offset in terms of reduced labour input and other resources required for insertion of an IUD. An IUD will result in contraception for as long as it remains in situ, often between five and ten years. c. A day case admission for an early stage medically treated termination costs £502, and from £514 to £649 for day case surgical terminations. Any admissions that are not dealt with as day cases will incur additional costs.
Use of ulipristal will result in less tangible costs arising from the need for a
prescription. For example, patients may have to book an appointment to see a
general practitioner or a nurse prescriber, whereas currently levonorgestrel can
be accessed from several sources without appointment.
Stated prices do not include VAT. Drugs and appliances that are not prescribed
on a primary care prescription will be subject to VAT, which will be 17.5% from 1st
January 2010.

Practical considerations

The availability of a PCC pill that can be taken up to 120 hours post-coitus and
which does not appear to demonstrate declining efficacy within that period could
have subtle and unintended consequences on patient and prescriber behaviour.
Levonorgestrel is widely available from a range of different sources. Many of
these sources of supply will not initially be available with ulipristal. This could
result in increased demand for medical consultations.
The product data sheet advises “[if experiencing a] delay of more than 7 days in
next menstrual period, abnormal bleeding at the expected date of menses, or
symptoms of pregnancy, pregnancy should be excluded by a pregnancy test.” As
these effects are relatively common following use of ulipristal (~20%) this could
result in a large number of patients seeking pregnancy tests, with the cost
potentially falling on NHS budgets. Additionally, there will be psychological
effects such as worry, panic, and anxiety, associated with any delayed
menstruation.
As with levonorgestrel, nausea and vomiting are relatively common adverse
effects with ulipristal. If patients vomit within three hours of taking ulipristal an
additional dose is required to ensure therapeutic efficacy is maintained. 4
North East Treatment Advisory Group. September 2009 Points to consider

• Ulipristal represents a new oral PCC option that is licensed for use up to 120 hours post-coitus. Evidence indicates that it is at least as effective as levonorgestrel and does not appear to demonstrate reduced efficacy with time. It has a similar adverse effect to levonorgestrel. • The evidence for the efficacy of ulipristal is derived from a population aged between 18 and 35 years. Use outside of this age group would not be evidence-based, although there is little reason to expect the efficacy to be different. • Ulipristal is about three times as expensive, per dose, than levonorgestrel, and about 1.5 to 2 times as expensive as an IUD. • Levonorgestrel is widely accessible via a number of established routes. Ulipristal would initially be available only via prescription. • The effects of inadvertent in utero exposure to ulipristal are not known.


Disclosure
The author has no relevant conflicts of interest to declare.
North East Treatment Advisory Group. September 2009 References
1. Guillebaud J. Contraception: Your questions answered. Churchill Livingstone, London, 2008
2. Wilcox AJ, Weinberg CR, Baird DD. Post-ovulatory ageing of the human oocyte and embryo failure. Human Reproduction 1998;13(2):394-7 3. European Medicines Agency. CHMP Assessment Report for Ellaone. Document reference EMEA/261787/2009 4. Summary of product characteristics. Ellaone (HRA Pharma). Accessed via the European Medicines Agency www.emea.europa.eu 5. Chabber-Buffet N, Meduri G, Bouchard P, Spitz IM. Selective progesterone receptor modulators and progesterone antagonists: mechanisms of action and clinical applications. Human Reproduction Update 2005;11(3):293-307 6. Blithe DL, Nieman LK, Blye Rp et al. Development of the selective progesterone receptor modulator CDB-2914 for clinical indications. Steroids 2003;68:1013-7 7. Gainer EE, Ulmann A. Pharmacologic properties of CDB(VA)-2914. Steroids 2003;68:1005-11 8. Creinin MD, Schlaff W, Archer DF et al. Progesterone receptor modulator for emergency contraception. Obstetrics & Gynecology 2006;108(5):1089-97 9. Cameron S, Glasier A, Fine P, Mathe H, Gainer E. Ulipristal acetate compared to levonorgestrel for emergency contraception within five days of unprotected intercourse: A randomised controlled trial. Abstract presented at the 8th Congress of the European Society of Gynecology, September 2009 10. Unpublished data on file courtesy of HRA Pharma, August 2009 11. Ulmann A, Scherrer B, Mathe H, Gainer E. Meta-analysis of emergency contraception trials comparing ulipristal acetate with levonorgestrel. Abstract presented at the 8th Congress of the European Society of Gynecology, September 2009. 12. Chabbert-Buffet N, Pintiaux-Kairis A, Bouchard P. Effects of the progesterone receptor modulator VA2914 in a continuous low dose on the hypothalamic-pituitary-ovarian axis and endometrium in normal women: A prospective, randomized, placebo-controlled trial. Journal of Endocinology and Metabolism 2007;92(9):3582-9 13. NHS dictionary of medicines and devices www.dmd.nhs.uk 14. Data provided by the Regional Drug & Therapeutics Centre (Newcastle) www.nyrdtc.nhs.uk 15. Department of Health. Payment by results tariff 2009-10 www.dh.gov.uk North East Treatment Advisory Group. September 2009

Source: http://ulipristal.s3.amazonaws.com/reviews/NETAG%202009.pdf