Contraception and HRT prescribing in patients on GLP-1 agonist weight-loss drugs

GP with an interest in women’s health Dr Toni Hazell discusses the recent rise in the use of injectable GLP-1 receptor agonists for weight loss and the challenges this can pose when prescribing contraception or HRT

GLP-1 agonists aren’t widely available on the NHS, so why is this a big issue?

The world of weight loss has been transformed by the arrival of glucagon-like peptide-1 (GLP-1) agonists (e.g. semaglutide), and the dual GLP-1/glucose-dependent insulinotropic polypeptide (GIP) agonists such as tirzepatide. For brevity, we will use the term ‘GLP-1 agonist’ to discuss both classes of drug. NICE guidance recommends tirzepatide at a BMI ≥35 kg/m² with one weight-related co-morbidity, but the NHS in England is not likely to meet that goal for another 12 years, with a much more restrictive initial criteria of a BMI ≥40 kg/m² and four out of a list of five co-morbidities.

GLP-1 agonists are widely prescribed in the private sector, usually with no face-to-face interaction, followed by a brief letter to the GP to inform them that their patient is taking this medication. Whilst the medicolegal responsibility for the GLP-1 rests with the prescriber, once we know that the patient is taking it, we would be liable if we continued to prescribe other medicines that interact, without taking that into account. Hormone replacement therapy (HRT) and contraception are commonly prescribed in primary care, and both are potentially affected by GLP-1 agonists. Adding the GLP-1 prescription to the computer screen as hospital issue is therefore sensible, as this will pick up possible interactions. Some practices have chosen to send a text message to any woman taking a GLP-1 asking them to arrange a review if they are taking oral contraception, as it may need to be changed.

What are the risks for someone using a GLP-1 agonist and oral contraception?

We know that weight loss can improve fertility – it is often recommended either to meet the criteria for fertility clinic referral, or to improve the chances of natural or assisted conception being successful. However there is a specific issue for women who are taking tirzepatide and who are relying on oral contraception to prevent pregnancy.

These are new drugs and we don’t therefore have a large amount of clinical data to look at, but there are pharmacokinetic studies which raise concerns. Tirzepatide (but not semaglutide) has been shown to have the following effects on oral contraception, which were not seen for semaglutide, or any of the less commonly used GLP-1 agonists. These will all act to reduce contraceptive efficacy.

Reduce the area under the curve (AUC) which represents the total drug exposure over time.
Increase the Tmax (the time taken to reach maximum plasma concentration).
Reduce the Cmax (peak concentration of the drug in the bloodstream).

It is always particularly concerning if a woman gets pregnant whilst taking a new drug, where we don’t have years of case reports about use in pregnancy. The high molecular mass of GLP-1s suggests that transfer across the placenta wouldn’t be expected, but given the lack of knowledge, women using them for weight loss (who do not have diabetes) are advised to stop before trying to conceive. Those using them for diabetes should discuss all of their medication with their clinician before trying to conceive.

In light of these findings, the College for Sexual and Reproductive Health (CoSRH) advises that women switch to a non-oral or barrier method of contraception for four weeks after initiation of tirzepatide, and four weeks after any dose change. After four weeks, some tolerance builds up, so that the slowdown of gastric emptying and reduction in absorption isn’t as pronounced.

A washout period should also be observed between taking the last dose of a GLP-1 agonist and trying to conceive – this is one month for tirzepatide, two months for semaglutide and 12 weeks for exenatide. They also should not be used when breastfeeding, again due to a lack of safety data rather than any positive data about harm.

What are the issues around GLP-1 agonists and HRT?

HRT is widely prescribed in primary care – women with obesity are most likely to be using a transdermal (TD) form of oestrogen, which will not be affected by a GLP-1 agonist. TD oestrogen does not increase the risk of venous thromboembolism (VTE) and is therefore recommended for all women with an increased background risk of VTE, including a BMI ≥30 kg/m². Women who have not had a hysterectomy need to combine the oestrogen in HRT with a progestogen; unopposed oestrogen carries a significant risk of endometrial hyperplasia and cancer, both of which are conditions for which obesity is also an independent risk factor.

Women taking a combined patch (or TD oestrogen with a levonorgestrel intrauterine device as their progestogen) are unaffected by the use of a GLP-1 agonist, as the entire regime is non-oral; the same would apply to a woman using a combined oral preparation, as any reduction in efficacy would apply equally to the oestrogen and the progestogen and so not increase the risk of endometrial proliferation. It is possible that the latter group might find a reduction in the efficacy of their HRT when using tirzepatide (for the same pharmacokinetic reasons discussed regarding contraception), but the likelihood and magnitude of this is unknown.

The group at risk of endometrial hyperplasia are those women using a GLP-1 and also using TD oestrogen (which will be unaffected by the GLP-1 agonist) and an oral progestogen, as the absorption of the progestogen may be reduced while the oestrogen stays the same, leaving the woman with an unbalanced regime and an insufficient amount of progestogen to protect her endometrium from the oestrogen.

The British Menopause Society (BMS) has made the following recommendations for this cohort, emphasising that data is notable by its absence:

Women using a GLP-1 should ideally use a non-oral form of progestogen alongside their TD oestrogen.
Where they prefer to continue with an oral progestogen, they should be informed of the lack of data to inform any dose adjustment.
A potential approach is to increase the dose of the oral progestogen for four weeks after starting a GLP-1 agonist, and after every dose increase until a stable dose is reached.
In the event of unscheduled bleeding, it may be wise to investigate earlier than suggested in national guidance for women using a GLP-1 agonist.

The BMS guidance did not say by how much the dose of oral progestogen should be increased – a document published by the Primary Care Women’s Health Society (PCWHS) suggested that a pragmatic approach would be to use the same increases as we would use for women taking a higher than usual dose of oestrogen, but also pointed out that the lack of data to guide this and emphasised that the levonorgestrel intrauterine device gives the best endometrial protection and should be recommended in this situation.

Sources and further reading