Eight million people in developing countries are on HIV treatment today,¹ but the number of people who need access to HIV medicines is growing.

© Brendan Bannon

In June 2011, governments committed to a target of reaching 15 million people with antiretroviral treatment (ART) by 2015 at the UN High-level Meeting on HIV/AIDS.² If this target is met, UNAIDS estimates that up to twelve million infections and more than seven million deaths can be averted by 2020, and that the number of new infections could be reduced by more than half by 2015.³

Now is a critical time to ensure widest possible access to ART: a landmark scientific breakthrough in 2011 showed that treatment with antiretroviral medicines (ARVs) not only saves lives, but can also stop HIV from being transmitted by up to 96%.⁴ In response to this evidence, in 2012 the World Health Organization (WHO) issued guidance supporting immediate treatment – regardless of a person’s immune system’s status or CD4 count – for HIV-positive people who have HIV-negative partners, in order to help prevent transmission of the virus.⁵ At the same time, guidance was also issued suggesting the possibility of offering full antiretroviral therapy for life to all pregnant women living with HIV for prevention of mother-to-child transmission of the virus (PMTCT). This protocol – ‘option B+’ – is easier to manage than starting and stopping PMTCT⁶ with each pregnancy, and better for mothers and their babies.

It is therefore critical to ensure the most robust and easiest-to-use medicines needed are made affordable to the largest number of people in developing countries. Over the past year, since the last edition of Untangling The Web, there have been mild price reductions for the newer ARVs recommended by WHO in its 2010 guidelines for first-line therapy, like tenofovir. Greater reductions are needed to support further scale-up. Meanwhile, second- and third-line ARVs remain priced many times higher than first-line drugs.

Further, middle-income countries are no longer offered standardised price discounts by originator companies, forcing them to negotiate ARV price reductions on a case-by-case basis. Some ground has been lost, as middle-income countries are once again paying exorbitant prices for the newest drugs. And battles over intellectual property in the world’s main producer of affordable ARVs – India – continue to heat up, while voluntary measures by companies to increase access require much closer scrutiny and investigation.

New producers on horizon expected to bring first-line prices down

Providing people with well-tolerated medicines with few side-effects that make it easier to adhere to treatment is one of the key ingredients to successful HIV treatment outcomes. This helps delay the need to switch to much more expensive second-line regimens or beyond, because of resistance.

Since 2006, WHO has recommended in its HIV/AIDS treatment guidelines that treatment providers begin moving away from the drug stavudine (d4T) because of its long-term irreversible side effects, and instead begin using tenofovir (TDF) or zidovudine (AZT).⁷ This call was reinforced in the latest guidelines released by WHO in 2010, with a clear recommendation to phase out d4T. In February 2011, the European Medicines Agency recommended that, in view of its long-term toxicities, d4T be used for as short a time as possible and only when no appropriate alternatives exist.⁸

Many developing countries have begun to use either TDF or AZT for new patients starting treatment. In a survey of 16 countries where MSF works, all had removed d4T from their national protocol as the preferred first-line ARV, with eight having chosen TDF, five AZT and three choosing either of the two as the alternative.⁹

The higher cost of these alternatives to d4T has however slowed this switch. Tracking the price evolution of TDF- and AZT-based first-line regimens is therefore critical.

Graphs 1 and 2 chart the prices of different first-line regimens today, and their evolution over time. They reveal:

• That the price of TDF-based regimens is now nearly the same (when combined with nevirapine), or lower (when combined with efavirenz), than AZT-based regimens, for countries that can access generic versions because patents do not form a barrier, or where voluntary licences allow access to generic versions. The price of TDF-based first-line regimens has in fact been decreasing since 2007 for those countries.

• That for countries that are unable to access the generics because of patent barriers or because they are excluded from the scope of voluntary licences, the prices remain consistently high. The fixed-dose combination of TDF/FTC/EFV (produced by Merck/BMS/Gilead) – which is an adherence-friendly one pill a day – has remained priced at $613 and $1033 per patient per year (ppy) for lower-income and lower-middle-income countries, respectively, for the last five years. For middle-income countries, prices can be even higher, as most companies have eliminated their standardised discount programmes for these countries, in favour of case-by-case price negotiations.

• That where patents are not a barrier, countries can make use of therapeutically equivalent alternatives to TDF/FTC/EFV which are considerably more affordable. For the past three years, a WHO-prequalified once-a-day combination of tenofovir/lamivudine/efavirenz (TDF/3TC/EFV) made by Matrix (Mylan) has been available. While its price has remained virtually static for the last year because it is the only product with WHO prequalification status, this is expected to change with new producers entering the market over the next year. This competition should help break Mylan’s monopoly on the market and bring prices down for this desirable combination. This in turn should help countries move away from d4T entirely.

• That until there these new fixed-dose versions become available, the same regimen can also be procured and administered as two separate pills, to be taken once a day (a double fixed-dose combination of TDF/3TC plus EFV). The price of this co-pack has come down by 20% since last year, to $113 ppy, making it the most affordable option of the WHO-recommended first-line regimens, with the added benefit of once-daily dosing.

In addition to being more affordable, TDF/3TC/EFV regimens have an advantage over AZT-based regimens thanks to their once-daily dosing (versus twice-daily for AZT) – see table 1. In addition, the latest WHO recommendations concerning EFV use during pregnancy are an additional incentive for countries to use this regimen – see box.

Improved option for second-line, but newer drugs remain expensive

With growing numbers of people in developing countries having been on treatment for a decade or even longer, ensuring the effectiveness of treatment, and long-term survival, depends on continuous access to newer and more potent ARVs. People need to switch to second-line regimens when they develop drug resistance over time.

© Bithin Das

In one of MSF’s longest-running HIV/AIDS treatment programmes, in Khayelitsha, South Africa, 12.2% of patients on treatment for five years needed to switch to a second-line drug combination because of virological failure. Demand for second-line treatment is growing fast: it is estimated that almost half a million people will need second-line medicines by 2012.¹²

A key development for second-line therapy over the past year was the tentative approval granted by the US Food and Drug Administration (US FDA) of the first-ever FDC of atazanavir/ritonavir (ATV/r) in November 2011. This FDC finally provides an alternative boosted protease inhibitor to lopinavir/ritonavir (LPV/r).

Because of the improved dosing of ATV/r – just one pill, once a day, compared to LPV/r with two pills, twice a day – it will likely become more prominent in second-line therapy. It is also more affordable than the lowest reported price for LPV/r.

However, it is crucial that additional suppliers enter the market to ensure global supply does not rely solely on one manufacturer.

Today’s most affordable second-line regimen (AZT/3TC + ATV/r) is priced at $399 ppy, down from $442 for last year’s most affordable combination. This however is still three times more than the most affordable first-line regimen. For countries where generic versions cannot be used because of patent barriers or because they are excluded from the geographical scope of the voluntary licences, the price can be many times higher.

For people who need to be switched from second-line therapy, a possible regimen of raltegravir + darunavir + ritonavir + etravirine is priced at $2,486 ppy at company discounted prices for sub-Saharan Africa and least-developed countries. This is over 14 times more than the most affordable first-line, and six times more than the most affordable second-line regimen (see graph 3).

Newer ARVs remain prohibitively expensive, partially because demand for some of them is still small. This is unlikely to change since patents have prevented the broad and open generic competition that will drive prices down. This means that in middle-income countries, where companies no longer offer standardised price discounts and prices must be negotiated on a case-by-case basis, much higher prices are paid (see table 2). In Russia, which is classified by the World Bank as an upper middle-income country, raltegravir, darunavir and etravirine combined was procured at over $27,000 ppy (see table 3). This does not even include ritonavir, which would be needed in addition to complete the regimen.

For those people already failing on their second-line combination, this unaffordable price will mean they almost certainly will be left without effective treatment options.

Looking into the pipeline: The best treatment options to support scale-up

Looking beyond today’s options, MSF, with the support of experts, outlined a vision of how to make best use of the ARVs in the near and more distant future in resource-limited settings, given the need to decentralise and simplify care even further to support treatment scale-up.

There are promising new drugs for the treatment of HIV in the future, including new classes of drugs that have new ways of preventing the virus from replicating. Some have the potential to be administered as long-acting formulations that would allow once-weekly or once-monthly dosing. Others could be potentially cheaper than the ARVs most commonly used today. And others still are currently used only for salvage therapy in developing countries – but already recommended for first-line therapy in some developed countries.

A meeting of experts on ARV sequencing¹⁷ convened by MSF in September 2011, together with Esther (Ensemble pour une solidarité thérapeutique hospitalière en réseau) and SOLTHIS (Solidarité thérapeutique & initiatives contre le SIDA), considered the following regimens for resource-limited settings in the short-, medium- and long-term. But ensuring access to promising drugs in the development pipeline will require surmounting patent barriers that prevent access.

Target characteristics of future treatment regimens

Delivering ART at scale in resource-limited settings requires more than efficacious drugs; six additional key principles guide ART choice. They are:

• Simplicity: Since high HIV-burden countries face a critical shortage of health workers, many countries have simplified HIV care to the point where it can be delivered by nurses at health centres. Keeping ART simple means prioritising FDCs, once-daily formulations, and regimens with minimal laboratory monitoring requirements.¹⁸

• Tolerability: Side effects are a major driver of treatment interruption, drug substitution, and treatment discontinuation, all of which undermine treatment and prevention efforts.^{19, 20} In particular, providing ART as prevention implies giving ART to people who may not yet have experienced clinical illness and may therefore be more reluctant to adhere to drugs with side effects.

• Durability: Regimens that have a high genetic barrier to resistance and include molecules with long half-lives will help reduce the risk of resistance associated with widespread, long-term ART delivery.

• Heat stability: Temperatures in Africa regularly exceed 30°C, and access to refrigeration is often limited (especially in rural settings). Some second-line regimens with boosted protease inhibitors must be kept at 2 – 8°C, which is a significant constraint.

• Universal applicability: Current regimens require frequent substitutions according to age, pregnancy, and interactions with other (non-HIV) drugs. The ideal regimen would be safe and effective irrespective of disease stage, usable throughout pregnancy, could be given to both children and adults, and be taken together with drugs for common co-infections, notably tuberculosis.²¹ This would eliminate the need for these substitutions.

• Affordability: Strategies that lower treatment costs should be prioritised; they include dose reduction, improved drug bioavailability, active pharmaceutical (API) cost reduction through improved chemistry process, and novel drug delivery systems and models of care. Strategies to remove barriers to generic competition should also be pursued.

© Brendan Bannon

Policies to bring prices down

Company-led discount programmes, based on the ‘differential prices’ that are mapped in this yearly report, have proven to be minimally effective at driving down the price of HIV medicines.²²It is the absence of patent barriers in key producing countries – or their removal – which enables generic competition to act as the main catalyst for lower prices.

Competition among multiple generic pharmaceutical manufacturers in countries where medicines were not patented, especially India, is what brought the cost of HIV/AIDS treatment down by 99% since 2000 (see graph 4). The lack of pharmaceutical patents in India until 2005 additionally allowed for the production of FDCs, which both support patient adherence and are crucial to the simplification of treatment that has been central to global treatment scale-up. India has thus been called the ‘pharmacy of the developing world’: more than 80% of donor-funded purchases of ARVs for use in developing countries from 2003 to 2008 were manufactured in India, and more than 80% of the ARVs MSF uses are sourced from India.²³

© Bithin Das

But increased drug patenting in key production countries, especially India, is starting to block price-busting generic competition for newer ARVs. If increased patenting and other intellectual property measures mean that generic competition cannot act as a catalyst to bring down the prices of medicines, tomorrow’s battle for access to affordable ARVs will require more systematic use of compulsory licences and other policies to bring prices down, or through voluntary licencing arrangements that respond to public health needs.

Compulsory licences: opening the door for lower prices

When drugs are patented, and pharmaceutical companies fail to make them available and affordable to patients in developing countries, governments can make use of their right, under international trade laws, to issue compulsory licences (CLs) to allow for open and broad generic competition. CLs are one of the public health safeguards enshrined in the TRIPS agreement,²⁴ which allow a government to override a patent by issuing a licence to a third party to produce or import a drug. CLs have proven to bring prices down dramatically by opening up the market to competition and thereby increasing access.

A number of CLs have been issued for ARVs in recent years, resulting in substantially lower prices for the medicines concerned. A compulsory licence in 2007 in Thailand brought the price of lopinavir/ritonavir down by 75%; that same year Brazil overcame a patent on efavirenz, enabling the government to import a generic version from India at one third of the originator company price, and a compulsory licence in Ecuador in 2010 halved the cost of lopinavir/ritonavir to the public health system.

This year India issued its first CL – for the oncology drug sorafenib tosylate, marketed as Nexavar by Bayer – in March 2012, because the drug was considered unaffordable and had not been made available in sufficient quantities in the country.²⁵ The move will lead to a 97% price reduction for the drug. This sets an important precedent. In India, CLs can be issued when patented drugs are marketed at an unaffordable price, as is the case for several of the newest ARVs, such as raltegravir and etravirine. This can potentially benefit not only Indian patients but also people in other countries if governments place orders with the generic manufacturer and so operationalise the CL provision for exports.

China too has lately adopted detailed procedural measures for the issuance of compulsory licences, in response to the potential need to use this flexibility to ensure access to medicines for the ongoing implementation of a universal healthcare programme, at a time when the country is unable to access more affordable versions of certain drugs as it is excluded from all recent voluntary licences (VLs).

Voluntary licences: read the fine print, if it’s public

If they have the right terms and geographic reach, VLs can enable generic competition and allow for the sale of generic products in developing countries where patents are in force, improving the affordability and accessibility of medicines. In issuing a VL, a patent holder authorises a generic manufacturer to produce and export a generic version of a medicine, often in exchange for royalty payments.

A number of ARVs are now being produced under different forms of licencing agreements between originator and generic producers. VLs with more than fifteen generic companies, primarily based in India or South Africa, have already been agreed with pharmaceutical companies Merck, Bristol-Myers Squibb, ViiV, Johnson & Johnson and Gilead for 21 key second- and third-line ARV formulations, and more are under negotiation. Annex 3 provides an overview of these different agreements.

The claims as to the public health benefits of these VLs can not be verified, as the contractual terms and conditions are not public. A number of different concerns have emerged, however, from the little information that is publicly available:

• There is no VL that covers all developing countries, so VLs do not address the problem of affordability of medicines in excluded countries. Originator companies tend to design these agreements to strengthen their control over key developing country markets. Generic manufacturers are only allowed to sell their more affordable versions in a limited list of countries, usually least-developed countries and sub-Saharan Africa only. This leaves out lower middle and middle-income countries (with the notable exception of South Africa and India, which are often covered). VLs thereby often limit the number of countries that can benefit from the finished products, regularly excluding countries with sizeable populations of people living with HIV.

• Likewise, all VLs have restrictions on which countries or manufacturers can produce the finished product. This excludes other countries with generic production capacity and limits competition.

• There are other clauses in VLs that can limit the open competition that can bring prices down, such as the limits on which countries and which sources can be used to obtain active pharmaceutical ingredients (API). This is a harmful limitation as the API is the most expensive component of a medicine.

• VLs often contractually bar generic producers from supplying countries in which the originator company’s patent applications are pending, leading to de facto monopolies – irrespective of whether a patent has been granted or not.

• Contractual restrictions on the supply to certain territories can also be used to block the use of a compulsory licence, as if a generic licencee were to supply excluded countries that had issued a CL, this would be a breach of the VL contract. Generic suppliers are therefore unlikely to take up the possibility of supplying to a country that had removed patent barriers through a CL, for fear that they would lose their VL contract if they fulfilled such a request.

There is one notable exception to this secrecy however. As an alternative to bilateral licencing, the Medicines Patent Pool (MPP), established in 2010, aims to develop voluntary licences transparently to manage patents collectively and in the interests of public health. The MPP could help overcome intellectual property barriers to facilitate the production for cheaper medicines, and allow the production of needed fixed-dose combinations that would otherwise require lengthy negotiations with numerous different patent holders, with an increasing impact if resulting licences cover all developing countries where patents are in force.

The MPP signed its first licencing agreement with a pharmaceutical company, Gilead, in July 2011, covering four drugs – tenofovir, emtricitabine, cobiscistat and elvitegravir – and combinations thereof. The licence slightly increases the geographic scope, from 95 to 112 territories, of an earlier licencing agreement on tenofovir signed bilaterally with generic companies and Gilead in 2006, but it continues to exclude key developing countries and restricts production to India.

However, there are a number of improvements. For example, the terms of the licence are public, allowing a full comprehension and public scrutiny of the conditions, which is an important step for transparency. It contains an explicit condition that the licence will not block a licencee’s right to supply countries excluded by the licence, should one of those countries issue a compulsory licence to override patents. The licence also provides a termination clause, so if the patents end up being rejected or not granted, a company that has signed the licence can terminate the agreement and freely manufacture the drug on its own.²⁶ This termination clause has already been used by two of the original 2006 licencees to terminate their licences on TDF.

Several VLs have been signed between originator companies and generic producers in parallel to the MPP, these bi-lateral deals have negotiated less favourable terms.

© Claudio Tommasini

What policies to rein in drug costs?

Supporting policies that can ensure newer and better medicines are made affordable for people in developing countries is a political choice – one that countries have committed to at the UN since 2001 and re-iterated in the Declaration that concluded the 2011 UN High Level Meeting on HIV/AIDS.

• Developing countries should make use of public health safeguards and other legal flexibilities to bring down the cost of medicines.

Least-developed countries should use their right not to grant or enforce medicines patents until 2016, and the World Trade Organization should extend this deadline beyond 2016.

Governments should routinely exercise their right to override patents in the interest of public health by issuing compulsory licences to bring down drug costs.

Countries should design flexible patent laws that favour access to medicines. They should have high standards for patentability to ensure only innovative products are rewarded with patent monopolies. India’s patent law for example contains key health safeguards, reserving monopoly status only for drugs that show a therapeutic benefit over existing ones. This prevents ‘evergreening’, a patent-extension strategy whereby a company makes a small change to a drug and receives a new 20-year patent. But this system is under threat – this part of law is under attack by pharmaceutical company Novartis, whose ongoing case against the Indian government is scheduled to be heard in the country’s Supreme Court in July 2012. A win for the company would likely lead to much broader patenting of drugs in India.

Patent oppositions are another crucial safeguard for access to medicines, so that undeserved patent applications can be challenged. India’s law allows any interested party to oppose a patent both before it is granted (through a ‘pre-grant’ opposition), or after (through a ‘post-grant’ opposition). For more about patent oppositions, visit www.patentoppositions.org

The use of these safeguards in the Indian law has resulted in the withdrawal of the patent applications on lamivudine/zidovudine and the rejection of key patent applications on tenofovir, darunavir, nevirapine syrup and lopinavir/ritonavir allowing generic companies in India to continue to manufacture, supply and export these HIV medicines to other developing countries.

In Brazil, the patent for lopinavir/ritonavir was annulled in February 2012 by the federal court of Rio de Janeiro, following an opposition filed by a manufacturer. This allowed for local production or importation of more affordable versions of a drug that the originator company Abbott was selling in Brazil for 47% more than the lowest global generic price. Oppositions have been filed on additional patents related to LPV/r.

South Africa, provides an example of where introducing all key public health flexibilities into the patent law could help promote public health and ensure generic competition by preventing excessive patenting and evergreening. In 2008 alone, South Africa granted 2,442 pharmaceutical patents, compared to only 278 in Brazil for the five-year period from 2003 – 2008.²⁷ A stronger patentability criteria and patent examination system in Brazil account for this difference. The Treatment Action Campaign and MSF have called on South Africa to amend its patent law to include stricter patentability criteria, a patent review system, pre- and post-grant opposition, and an administrative rather than judicial process for granting compulsory licences, in the interest of public health.

Flexibilities that are proven to be ineffective should be reformed. World Trade Organization members should review and reform the August 30 Decision, designed to allow the exporting and importing of medicines under a compulsory licence to countries which have no manufacturing capacity and cannot produce their own generic medicines.

The experience of the only use of this flexibility in Canada has shown that what was intended to be an ‘expeditious solution’ fails to meet the needs of developing countries with no or insufficient manufacturing capacity to import medicines patented in drug-producing countries.²⁸

• Developed countries should refrain from pushing measures that impose even greater intellectual property protection than international trade rules require.

A number of multilateral trade deals currently exceed the TRIPS agreement in their intellectual property protection requirements. The Trans-Pacific Partnership Agreement currently under negotiation between the US and Pacific Rim nations seeks to impose high levels of intellectual property protection that will limit access to medicines and could become a dangerous template for future agreements.

The European Union and the European Free Trade Association countries are currently pushing for policies to be included in free trade agreements with India, such as enforcement and investment provisions that will further restrain competition and impact access to affordable medicines.

By pushing drug prices up, such agreements also directly undermine efforts by US and European donor governments to finance and support treatment scale-up.

The US’s Special 301 mechanism is a further example of a bilateral punitive measure to challenge efforts by developing countries to ensure access to medicines for their populations, and to drive countries to implement intellectual property measures into their domestic laws above those required by international trade law.

• All countries should refrain from introducing IP enforcement measures that limit the production, export, transit and importation of generic medicines.

Agreements and laws that conflate legitimate generic medicines with deliberately falsified medicines under the term ‘counterfeit’ should be avoided, reversed or rejected. The concerns over the effects on access to generic medicines of the Anti-Counterfeiting Trade Agreement (ACTA),⁴⁵ for example, have led to the EU Parliament rejecting the treaty.

The harm such provisions can do was recognised in the decision of the Kenyan High Court in 2012 when it struck down, as a breach of the constitutional right to life, human dignity and health, three sections of the Kenyan Anti-Counterfeiting Act 2008 that would have threatened access to generic medicines for people living with HIV. On going discussions at the East African Community level on anti-counterfeiting laws must ensure that similar provisions are not included.

The European Commission is currently reviewing the EU customs regulations that have in the past been used to detain generic medicines in transit through the EU on their way from one developing country to another. The proposed amendments tabled by the European Commission fail to ensure the smooth and free movement of legitimate generic drugs across trading routes and require further amendments.

• All companies should make the terms of all voluntary licences public, and support mechanisms that encourage transparent, collectively managed and public-health focused licencing, by participating in the MPP rather than seek to create alternative approaches.

Zimbabwe is moving ahead with its plans to start giving tenofovir to all patients on first-line antiretroviral treatment. The switch to tenofovir is happening in a phased manner due to financial limitations. Dr. Steven Van Den Broucke, who worked for MSF in the country from July 2010 to March 2012, explains the benefits for MSF and other health care providers of switching patients to tenofovir.

© Brendan Bannon

“We started to switch groups of patients who were initiated on stavudine to tenofovir last year, and now all patients are due to be switched to treatment with tenofovir by the end of this year.

A major reason for making the switch was that the drug is much easier to use in nurse-driven programmes. Using tenofovir in combination with efavirenz, for example, in a triple fixed-dose combination means it is possible now for us to initiate a patient on just one tablet, once a day, right from the very start of treatment. Also hugely beneficial is that fact that this combination of drugs – efavirenz, tenofovir and lamivudine – doesn’t interact with medicines the patient may be taking to treat tuberculosis at the same time.

The major problem with stavudine, which we are now replacing with tenofovir, has been the side effects it can produce in patients, such as lactic acidosis and peripheral neuropathy, both of which require careful medical monitoring.

Lactic acidosis is difficult to diagnose and while peripheral neuropathy can be spotted more easily, it is difficult to treat. And if it isn’t picked up early enough, the side effect becomes irreversible and can cause a lot of pain and prevent the patient from being able to walk. Both of these factors mean that it is much harder to manage a patient on stavudine than on tenofovir in a nurse-driven setting.

Also, because tenofovir causes fewer side effects in patients, it opens up the possibility, from a medical perspective, to increase the amount of time between patients’ clinic visits. Patients could in theory pick up a six-month supply of the drug when they visit the clinic, instead of the current usual three-month supply here. This would result in lower workload for programme staff. Since patient cohorts are always increasing, anything that can reduce workload within a programme is valuable.

After the green light from the national authorities, MSF started to provide tenofovir in combination with efavirenz, instead of with nevirapine – it’s a preferable combination because it means patients only have to take one pill a day. But it is more expensive. We are currently providing this treatment ourselves in our projects, but hope to see the price of the combination drug drop within the next couple of years, allowing for a wider take-up across Zimbabwe, beyond MSF projects.”