Treatment outcomes no better with single tablet regimens than individual tablets

One pill a day HIV treatments such as Atripla, Stribild and Eviplera have the same rates of virological failure, drug resistance and side-effects as multiple tablet regimens, according to a meta-analysis presented to the British HIV Association conference in Brighton today. Single tablets cost the NHS five times more but have unproven clinical benefits, Andrew Hill of the Chelsea & Westminster Hospital said.

Typically, new drugs stay on patent for 20 years. This means that the pharmaceutical company which originally developed the drug has the exclusive right to sell it. Following this, other manufacturers can produce generic versions of the drug, which are equivalent to the branded product, but are typically 80% cheaper.

The NHS has been encouraging the increased use of generic equivalents wherever possible across a wide variety of health conditions.

With the NHS currently spending £411 million a year on antiretroviral drugs alone and the number of patients expected to keep on growing, there is a particular need to control this spending. Generic versions of efavirenz, lamivudine, ritonavir, zidovudine and nevirapine are already available in the UK. Within the next three years, tenofovir, abacavir, lopinavir, atazanavir and darunavir will also come off patent.

However, pharmaceutical companies often use a strategy known as ‘evergreening’ in order to extend the period of time in which they have patented products in the market. One form this can take is the development of fixed-dose combinations and single tablet regimens.

For example, lamivudine (Epivir) was first produced in 1990. Although the patent on this expired in 2010, GlaxoSmithKline had by then produced a pill that combined lamivudine with zidovudine (Combivir) which remained under patent until 2013. They also combined lamivudine with abacavir in Kivexa, which they will have the exclusive right to manufacture until 2019. Furthermore, the company has recently launched a one-pill-a-day HIV treatment, Triumeq, which puts lamivudine, abacavir and dolutegravir into a single tablet regimen.

Another example of evergreening is Gilead’s tenofovir alafenamide fumarate (TAF), an alternative formulation of tenofovir that is safer for the kidneys and bones. While TAF was discovered in 2001 it will probably only be launched a few months before the original drug’s patent expires in 2017.

The development of single tablet regimens may meet a demand from people living with HIV and clinicians for treatments that are more convenient and easier to take, but can the high price tags be justified by better patient outcomes?

Antiretroviral regimens made up of generic drugs in three separate tablets typically cost less than £1000 per person per year. Branded single tablet regimens cost between £4500 and £7400 per person per year.

No clear benefit to co-formulated tablets

In order to find out whether there are any differences in clinical outcomes that could justify the increased costs, Andrew Hill and colleagues conducted a systematic review and meta-analysis, pooling the results of randomised trials which compared co-formulated antiretrovirals with individual components. Included are:

• Five older trials in which, for example, the fixed-dose combination Kivexa was compared with its individual components, abacavir and lamivudine. Participants needed to take an additional drug (a protease inhibitor or non-nucleoside reverse transcriptase inhibitor) separately and some trials involved twice-a-day regimens.
• Four more recent trials in which single tablet regimens such as Atripla, Stribild and Eviplera were compared with the use of regimens of three or four tablets.

A total of 2568 people took part. All the single tablet regimen trials involved people who were already stable on antiretroviral therapy and virally suppressed switching to a new regimen. The lack of randomised studies in people beginning HIV treatment for the first time is a limitation. But Andrew Hill said that if pharmaceutical companies believe that single tablet regimens really are better for new patients, it’s up to them to run the randomised controlled trials to demonstrate this.

Observational cohorts – which provide a lower standard of evidence and are subject to biases in patient selection – were not included.

The primary outcome of interest was virological failure (viral load above 400 copies/ml on two consecutive occasions). This happened as infrequently with individual drugs as with co-formulated pills, in both groups of trials. (The very slight reduction in risk of virological failure with combined pills was not statistically significant: 1.3%, 95% confidence interval -2.8% to +0.2%.)

Similarly, there was no significant difference in cases of drug resistance or of people changing treatment because of side-effects.

However, a slightly larger number of people joining trials of single tablet regimens but randomised to receive individual tablets changed their treatment despite having an undetectable viral load. Hill speculated that this might be explained by people wanting to join a trial in order to access single tablet regimens and then dropping out when randomised to the ‘wrong’ arm. (4.8% more switches, 95% confidence interval 1.6% to 7.9%.)

Data on patient preferences did show that co-formulated pills are the more popular option, even if they weren’t associated with any benefits as measured in Quality of Life scores. The early trials found 5% better adherence with fixed-dose combinations whereas the more recent trials’ adherence data are less clear. Moreover, any adherence benefits didn’t translate into actual differences in clinical outcomes.

Andrew Hill concluded that the benefit of fixed-dose combinations over individual generic pills has not been proven in randomised trials, despite the large differences in costs.

Rising antiretroviral costs

In a plenary talk yesterday, Dr Hill said that the high price of branded drugs cannot be justified by the cost of the raw materials and of production. Investigation of the prices charged by Indian and Chinese generic factories suggest very low costs.

For example, a year’s supply of the hepatitis B drug entecavir costs around £24 to produce and a generic version is available in India for around £285. Bristol-Myers Squibb sells branded versions in India for £875, in the UK and some other European markets for £4600, and in the United States for around £10,000.

Similarly, the new hepatitis C drug sofosbuvir costs £70 to produce. Prices are far higher in Western countries, but with substantial variations that cannot easily be explained – £18,000 in Spain, £35,000 in the UK and £57,000 in the USA.

While antiretrovirals are not as expensive as this, their prices rise every year by about 8%. Moreover, the number of people receiving HIV treatment in the UK rises every year at a similar rate.

This year 78,000 people are receiving antiretrovirals, with a cost to the NHS of £411 million. By 2019, we can expect 106,000 people to be on antiretrovirals, with the bill rising to £559 million. The numbers do not include people taking pre-exposure prophylaxis (PrEP) or hepatitis C treatment, although both need to be financed by the same specialised services budget of NHS England.

Over five years, the antiretroviral drug bill could be £2.41 billion if branded drugs, included some co-formulations, continue to be used. Switching whenever possible to generic versions of the same drugs would reduce the bill to £1.16 billion, a potential saving of £1.25 billion.

Daily pill counts would not be substantially different as appropriate single tablet regimens are not always available. They would rise from an average of 2.3 pills in the first scenario to 3.5 pills with the use of generics (or fewer if generic co-formulations became available).

Larger savings might be achieved if competition between generic suppliers leads to lower prices. In contrast, a lack of competition between suppliers or slow uptake by patients and clinicians would minimise the savings.

Andrew Hill concluded that companies which have invested in research and development should be rewarded with appropriate value-based prices for their new medicines, after evaluation by the National Institute for Health and Care Excellence (NICE). Patents for new medicines should be respected.

But evergreen patenting should be closely monitored, he said. When patents expire, medicines should be provided at generic prices close to the cost of production. When a generic drug becomes available, cost-effectiveness analyses and the prescribing decisions they inform must be re-evaluated for other drugs that treat the same condition. Existing prices may no longer be justified.

Patients should be systematically asked to switch to generic drugs when they become available – such a strategy could save the NHS £1.25 billion.

Switching strategies

The real-life issues involved in encouraging greater use of generics were reflected in several presentations at the conference.

Royal Liverpool University Hospital reported that although local guidance recommended generic efavirenz and Kivexa as the standard first-line therapy, only 8% of prescriptions complied with this. Doctors only occasionally recorded their justification for giving Atripla instead.

Clinicians in Middlesborough persuaded around half those taking Atripla to switch to generic efavirenz and Truvada. While most patients were satisfied with the change, up to a third were not.

Finally, when 96% of patients in Copenhagen were switched from Atripla, the drug company responded by dramatically lowering their prices – which meant that a year later most patients could switch back.

In all these settings, clinical outcomes after switching were excellent.