Availability and affordability of new medicines in Latin American countries where pivotal clinical trials were conducted

Abstract Objective To assess whether new pharmaceutical products approved by the United States Food and Drug Administration (FDA) in 2011 and 2012 were registered, commercialized and sold at affordable prices in the Latin American countries where they were tested. Methods We obtained a list of new molecular entities (new pharmaceutical products) approved by the FDA in 2011 and 2012. FDA medical reviews indicated the countries where pivotal clinical trials had been conducted. The registration status of the products was obtained from pharmaceutical registers; pharmaceutical companies confirmed their availability in national markets and local pricing observatories provided the price of medicines in retail pharmacies. Affordability was assessed as the cost of a course of treatment as a proportion of monthly income. Information on safety and efficacy was gathered from independent drug bulletins. Findings Of an expected 114 registrations, if the 33 products had been registered in all the countries where tested, only 68 (60%) were completed. Eight products were registered and commercialized in all countries but 10 had not been registered in any of the countries. With one exception, products for which we obtained pricing information (n = 18) cost more than the monthly minimum wage in all countries and 12 products cost at least five times the monthly minimum wage. Conclusion Many pharmaceutical products tested in Latin America are unavailable and/or unaffordable to most of the population. Ethical review committees should consider the local affordability and therapeutic relevance of new products as additional criteria for the approval of clinical trials. Finally, clinical trials have opportunity costs that need to be assessed.


Introduction
The high cost of many new medicines calls into question whether people in low-and middle-income countries will have access to them, 1 yet an increasing number of pivotal trials are carried out in these countries. Pivotal clinical trials are those included in the applications for market authorization or new drug approval documents submitted to regulatory agencies. This paper explores two issues: (i) are new molecular entities (hereafter products) approved by the FDA in 2011 and 2012 available in Latin American countries where the pivotal trials were conducted? and (ii) if registered, are they marketed at affordable prices? We also obtained information about the therapeutic relevance of the new products from the databases of two independent drug bulletins.

Methods
This is a cross-sectional study. We obtained the list of new products approved by the United States Food and Drug Administration (FDA) in 2011 and 2012. 2,3 One product, Gadobutrol (Gadovist), was approved during the study period but excluded from the study because it is a contrast dye used in radiology, not a pharmaceutical treatment. The FDA's medical reviews of the new products, included in their drug approval history, 4 provided the names of countries where the trials had been conducted. If this information was not available in the medical reviews, we obtained it from the trial sponsors.

Obtaining regulatory status
We searched the pharmaceutical registers for the regulatory status of relevant products in each country. The information included in the registers varies slightly by country. Brazil, Chile and Colombia maintain a register of approved pharmaceuticals; Argentina has a register of marketed products; Mexico publishes a list of the products approved per time period and Peru catalogues products available in pharmacies (See Box 1 for a list of web sites consulted). For the countries without registers (Costa Rica, Ecuador, Panama, Peru, Uruguay) we approached the regulatory agencies. We were unable to reach the regulators in the Dominican Republic or the Bolivarian Republic of Venezuela.
We contacted pharmaceutical companies' headquarters in the United States of America to gather information on the marketing status of their products in the selected countries.

Cost of medicines
In Latin America, about 78% of all medicines are paid for outof-pocket in retail pharmacies. 18 Since the products of interest were not included in the World Health Organization (WHO)/ Health Action International (HAI) medicine prices' database, we obtained the price of the unit dose of each product from the countries' price observatories, which report the maximum price to consumers (Brazil, Mexico) or the observed consumer prices (Argentina, Chile, Colombia, Ecuador, Peru; Box 1).
The consumer prices for Argentinean products not tracked by its observatory and for products in Costa Rica, where there is no observatory, were provided by pharmacological experts who obtained them from local distributors. The quantities needed to complete a course or a year of treatment were calculated using the FDA-approved product label. The pricing information was gathered between 25 August and 20 September, 2014.
Objective To assess whether new pharmaceutical products approved by the United States Food and Drug Administration (FDA) in 2011 and 2012 were registered, commercialized and sold at affordable prices in the Latin American countries where they were tested. Methods We obtained a list of new molecular entities (new pharmaceutical products) approved by the FDA in 2011 and 2012. FDA medical reviews indicated the countries where pivotal clinical trials had been conducted. The registration status of the products was obtained from pharmaceutical registers; pharmaceutical companies confirmed their availability in national markets and local pricing observatories provided the price of medicines in retail pharmacies. Affordability was assessed as the cost of a course of treatment as a proportion of monthly income. Information on safety and efficacy was gathered from independent drug bulletins. Findings Of an expected 114 registrations, if the 33 products had been registered in all the countries where tested, only 68 (60%) were completed. Eight products were registered and commercialized in all countries but 10 had not been registered in any of the countries. With one exception, products for which we obtained pricing information (n = 18) cost more than the monthly minimum wage in all countries and 12 products cost at least five times the monthly minimum wage. Conclusion Many pharmaceutical products tested in Latin America are unavailable and/or unaffordable to most of the population. Ethical review committees should consider the local affordability and therapeutic relevance of new products as additional criteria for the approval of clinical trials. Finally, clinical trials have opportunity costs that need to be assessed.

Commercialization status
Obtaining information on the commercialization status of each product from the pharmaceutical companies was difficult. The headquarters of some companies responded quickly with the requested information (Vertex, Exelixis, Glaxo Smith Kline (GSK), Bristol-Myers Squibb (BMS), Sanofi, AstraZeneca), while others referred us to their country subsidiaries or to the companies responsible for commercializing their products outside the United States of America (USA). The accuracy of the information provided depended on the familiarity of the respondent with the company's practices and databases. Two companies provided contradictory information and one referred us back and forth between the original producer and the licensee. With few exceptions (Pfizer Brazil, Colombia and Mexico; Janssen Argentina; Novartis Argentina and Colombia; Takeda Brazil and Boehringer Mexico), the Latin American offices were less willing to share information than the respondents at headquarters. One company representative wrote: "In response to your question below, we have a policy of restricting the disclosure of proprietary business information/strategies unless we have a formal business relationship protected by a confidential disclosure agreement. " Information on registration status and pricing helped resolve some of the inaccuracies of information reported by industry. For example, pricing information indicated the probable availability of pertuzumab in Mexico, rivaroxavan in Colombia and Mexico and ticagrelor in Argentina. We decided that bosutinib was not marketed in Argentina or Peru since it was not included on the list of marketed products (Argentina) or in the catalogue of products available in pharmacies (Peru) and because price information was not available in either country.
It was impossible to confirm the marketing status in 10 cases (pasireotide in Brazil; rilpivirine in Argentina, Chile and Mexico; pertuzumab in Peru; teriflunomide in Chile and Mexico; tofacitinib in Costa Rica and Peru; and vandetanib in Mexico).

Measuring affordability
There are challenges in gathering information related to the price of medicines and incomes to determine affordability thresholds. 19 Some argue that a threshold of 5% of total expenditure for the purchase of medicines would classify them as unaffordable in countries such as India and Indonesia. 20 Some au-thors consider health-care expenditure catastrophic if it exceeds 10% of yearly household income. 21 In this study, the affordability of each course of treatment is presented in relation to the monthly household or per capita income. Whether or not this is a catastrophic expenditure depends on the socioeconomic status of the individual or the household. In regions with large income inequalities, such as Latin America, using population averages can be misleading. Wealth was measured using: (i) monthly minimum wages in 2014, obtained from public announcements in the media; (ii) average monthly per capita income, from a World Bank database; (iii) the monthly household net adjusted disposable income, and (iv) the monthly household financial wealth which was only available for Brazil, Chile and Mexico (Table 1).
Finally, we consulted the databases of two independent drug bulletins (Prescrire, France [http://www.prescrire. org/fr/Search.aspx], and the Health Research Group -Public Citizen, USA [http://www.worstpills.org/search/]) for assessments of the clinical relevance of the new products compared with existing treatments.

Findings
The 33 products included in this study are shown in Table 2. Of an expected 114 registrations, if the 33 products had been registered in all the countries where tested, 60% (68/114) were completed. Three cases were excluded as the regulatory status of the product Combining information on registration and availability indicated that 30% (10/33) of products were not registered or commercialized in any of the countries where they had been tested (aclidium bromide, axitinib, bedaquiline, bosutinib, cabozantinib, elvitegravir/obicistat/emtricitabine/tenofovir disoproxil fumarate, lucinactant, perampanel, tbo-filgastrim, ziv-aflibercept); two products were registered but not marketed in any of the countries where tested (enzalutamide and ezogabine) and eight were registered and commercialized in all countries where tested (aflibercept, indacaterol maleate, ipilimumab, linagliptin, regorafenib, roflumilast, taliglucerase alfa, telaprevir). Table 3 presents the approval and commercialization status in September 2014 of the products included in the study that were not registered or commercialized in all the countries where tested. In two cases the pharmaceutical company stated that a product was available when the regulatory agency said it had not been approved and in several cases the products were registered but not marketed. In total, by September 2014, 36.4% of products were not marketed in any of the countries where they were tested and only 51% were available where tested.
The cost of one course of treatment (or a year of treatment for chronic conditions) is shown in Table 4. The prices of 18 of the 21 products marketed in Latin American countries for which we were able to obtain information varied widely. In all countries, and for all except one of the products, the cost was greater than the monthly minimum wage. Five products cost one to four times the monthly minimum wage, while six cost between 100 and 896 times this amount.
Three products in Argentina and Chile -and two products in Brazil -cost less than the average monthly per capita income (Table 5). Only one product (rivaroxavan) cost less than the average monthly per capita income in all countries, but four products cost more than 100 times this amount. The results using monthly household disposable income and monthly household wealth were very similar to findings for average monthly per capita income (details available from the corresponding author).

Clinical trials in Latin America Núria Homedes & Antonio Ugalde
The bulletins of Prescrire and/or the Health Research Group evaluated 25 of the 32 products and determined that 20 (80%) had no advantage over existing treatments and had significant side-effects. According to these sources, and other bulletins reviewed by Prescrire, the remaining five products (crizotinib, enzalutamide, ipilimumab, pasireotide, telaprevir) may provide some thera-peutic benefit to a subset of patients, but the risk-benefit ratio was uncertain (details available from the corresponding author).

Discussion
According to our data, 20 months after the FDA had approved the commercialization of 33 products in the USA, only eight (25%) had been registered and commercialized in all the Latin American countries where they had been tested. Thirty percent (10/33) had been neither registered nor commercialized in any of the countries; 45% (15/33) were registered and some of these were commercialized in several countries.
The suboptimal implementation of differential or tiered pricing for pharmaceuticals, a strategy recommended by WHO to enhance access to pharmaceuticals, 22  Within pharmaceutical companies, there seems to be little communication between the research and development units and those responsible for marketing the final products. Clinical trials are outsourced when the countries meet the conditions of the sponsor or of the contract research organizations managing the trial, such as having large urban centres with researchers willing to enrol patients, expedited approval of protocols and sufficient patients who can be readily recruited. 23 The registration and marketing of new products are business decisions based on a country's regulatory conditions, the presence of a business affiliate or partner, the willingness of the health system to include a product in its formulary, the number of patients who can afford the treatment and estimates of drug profitability for the company.
Latin American agencies responsible for the approval of clinical trials, including the research ethics committees, do not consider if the tested products will be available and affordable. Only Brazil requires that all drugs tested in the country be registered when found to be safe and effective. Resolution 446 of its regulatory agency (Anvisa) reads: "When developing new drugs, if safety and effectiveness is proven, its registration is obligatory in Brazil." However, as our findings demonstrate, it appears that the regulation is not being enforced.
In sum, neither the sponsors of the clinical trials, nor the regulatory agencies, nor any of the bodies that approved ethical declarations regarding clinical trials have suggested pre-trial mechanisms to ensure that new products will be available and affordable in those countries where testing has taken place. "In general, if there is good reason to believe that a product developed or knowledge generated by research is unlikely to be reasonably available to, or applied to the benefit of the population of a proposed host country or community after the conclusion of the research, it is unethical to conduct the research in the country or community."(Commentary on Guideline 10). 27 Latin American regulatory agencies have very limited resources for assessing the safety and efficacy of new products, some of which are very complex. For this reason, they depend on the FDA or the European Medicines Agency's decisions to grant approval. The pharmaceutical industry claims that conducting trials in Latin America strengthens research capacity in biomedical science, 23 but it could also be suggested that high payments to principal investigators lure some of them away from developing other products needed in the region, such as treatments for dengue, malaria Box 2. International ethical guidelines asserting that post trial access to treatment should be ensured

Council for International Organizations of Medical Sciences 24 Guideline 10: Research in populations and communities with limited resources
Before undertaking research in a population or community with limited resources, the sponsor and the investigator must make every effort to ensure that: • the research is responsive to the health needs and the priorities of the population or community in which it is to be carried out; and • any intervention or product developed, or knowledge generated, will be made reasonably available for the benefit of that population or community.

Commentary on Guideline 10
This is applicable especially to research conducted in countries where governments lack the resources to make such products or benefits widely available. Even when a product to be tested in a particular country is much cheaper than the standard treatment in some other countries, the government or individuals in that country may still be unable to afford it. If the knowledge gained from the research in such a country is used primarily for the benefit of populations that can afford the tested product, the research may rightly be characterized as exploitative and, therefore, unethical. The negotiation should cover the health-care infrastructure required for safe and rational use of the intervention, the likelihood of authorization for distribution and decisions regarding payments, royalties, subsidies, technology and intellectual property, as well as distribution costs, when this economic information is not proprietary. In general, if there is good reason to believe that a product developed or knowledge generated by research is unlikely to be reasonably available to, or applied to the benefit of, the population of a proposed host country or community after the conclusion of the research, it is unethical to conduct the research in that country or community.

Universal Declaration on Bioethics and Human Rights. Article 15 -Sharing of benefits 25
Benefits resulting from any scientific research and its applications should be shared with society as a whole and within the international community, in particular with developing countries.

Declaration of Helsinki (2013) 26
Item 22. In clinical trials, the protocol must also describe appropriate arrangements for posttrial provisions.

Clinical trials in Latin America Núria Homedes & Antonio Ugalde
and leishmaniasis or from the development of generic biophamaceuticals, a move that could save lives and money. 28,29 Industry prices are not related to product development cost 30,31 and some new products are unaffordable even in high-income countries. In both the USA and the United Kingdom of Great Britain and Northern Ireland, [32][33][34][35] physicians and health authorities are increasingly reluctant to accept expensive medications offering little improvement over existing cheaper alternatives with demonstrated safety and efficacy.
In January 2014, a new consensus framework for ethical collaboration between patients' organizations, healthcare professionals and the pharmaceutical industry was published. 36 The framework states that: "Continuing to advocate and support the principle that all human subject research must have legitimate scientific purpose, aims to improve health outcomes and be ethically conducted…" 36 To comply with this consensus and with universally-accepted ethical principles, 27,37,38 the pharmaceutical industry should: (i) reconsider research and commercial strategies to ensure that new products add therapeutic value to the existing therapeutic arsenal at an affordable price; (ii) include the estimated local prices for potential new products in clinical trial protocols so that the regulatory agencies and research ethics committees can consider the affordability of the product when authorizing the research (iii) work with regulatory agencies in the countries where products are tested to ensure registration and availability of new products that prove to be safe and effective.

Study limitations
The information on registration and commercialization status of new products may contain inaccuracies. We identified and corrected some errors through triangulation, but others may not have been detected. Some FDA reviews did not specify which of the clinical trials were pivotal. Even though we also gathered information from trial sponsors, we may have included trials that technically might not be considered pivotal. There are unavoidable limitations in estimating thresholds of affordability, since financial sacrifices and risks cannot be easily defined by others. These are personal decisions that are influenced by personal values and culture.
Evaluating the price of drugs continues to be complex and currently there is no standard method. Despite their shortcomings, country observatories, which tend to be based on the WHO/ HAI methods, are probably the best and most reliable sources of information. Currency variations add to the complexity of reporting pricing information across countries. We priced the drugs in September 2014, but the data used to determine monthly income is from 2013. Moreover, in the Latin American countries included in this study, income is very poorly distributed. If we were to remove the highest two income deciles, the income per capita for the rest of the population would be drastically reduced, and therefore the affordability threshold would have to be lowered.