Article Commentary: Generic Substitution of Antiretroviral Agents: Never a Problem?

More than 30 years after the first antiretroviral agents were introduced to treat HIV infection, many of these agents have now become available as generic formulations in Western countries. Initially, the first generics represented ‘older’ antiretroviral agents such as zidovudine and nevirapine. These agents are no longer the recommended first-line options according to DHHS [1] and IAS-USA guidelines [2]. More recently, the most frequently used nucleoside reverse transcriptase inhibitor backbone in Western countries, tenofovir disoproxil fumarate/emtricitabine (TDF/FTC), has become generic. In the Netherlands there are currently six different manufacturers, which have made it possible to reduce prices in the range of 80–90% compared with the original price of the innovator product (Truvada®). This reduction is partly possible by gaining a market share due to the pre-exposure prophylaxis (PrEP) indication.

The latest generic introductions in the Netherlands has been darunavir and ritonavir, and we expect the next generic introductions, for example raltegravir, in 2021. With the latest generic introductions it is now possible to provide triple therapy completely generic, for example TDF/FTC plus darunavir/ritonavir. Since health-care costs for HIV-infected patients are responsible for approximately 80% of total drug costs for HIV-infected patients, cost savings through generic drug substitution should be a priority in any national healthcare system [3]. Inclusion of generic medication in the existing guidelines would also help this transition. In addition, it would be helpful if we also spread the positive effects of generic medication, thereby increasing confidence among physicians, pharmacists and patients.

Generic substitution is not unique for HIV treatment but in contrast to other disease areas (for example, hypertension, hyperlipidaemia, etc) it is a relatively new phenomenon for HIV-infected patients and possibly for their health-care providers too. Therefore, we think that generic substitution in HIV treatment is more novel than for patients with hypertension or hypercholesterolaemia, and moreover it follows similar patterns of more sustained preferences for innovator products as seen in patients treated with drugs with narrow therapeutic ranges, for instance patients with organ transplants or epilepsy. We have various reasons for holding the opinion that generics in HIV are special, and we want to explain this here.

The first aspect that needs to be discussed is that the bioequivalence criteria for licensing a generic formulation of an antiretroviral drug are similar as for any other drug. That is, the 90% CI of the geometric mean ratio of area under the curve (AUC) and maximum concentration (C_max) of the antiretroviral drug in the generic formulation versus the innovator product should lie within 80–125%. Regulatory authorities have adapted this criterion for drugs with a narrow therapeutic range to 90–111%, and research has demonstrated that licensed generic antiretrovirals to a large extent also meet this more strict criterion [4]. Some physicians argue that this interval is too wide for HIV drugs and that individual differences within a patient may be larger than 20%. Although theoretically this argument is valid, we think there are two arguments against this statement: first, current antiretroviral agents that become generic are much more forgiving and relatively small differences in exposure will not put the patient at increased risk for therapy failure [5]; second, health-care providers and patients generally do not realize that also intrapatient variability exists with innovator products, so variability has always been part of HIV treatment, also before generic formulations arrived. Research with the antiepileptic agent gabapentin demonstrated that intra-patient variability because of generic substitution cannot be distinguished from intrapatient variability while continuing taking the innovator product [6].

In addition to the bioequivalence demonstrated for any licensed generic antiretroviral agent, one should also realize that millions of HIV-infected patients are currently being treated with complete generic ART combination, namely outside Western countries. Although results from cohorts in resource-limited countries cannot directly be compared with cohorts in Western countries, there is no indication of suboptimal treatment because of the use of generics. It is a good thing that investigators from the HIV-NAT 223 study [7] have presented data that generic switch of atazanavir/ritonavir showed equivalent pharmacokinetics, safety and efficacy. It is understandable that such studies will not be carried out for every particular generic formulation, but the team in Thailand needs to be complimented that they have undertaken this study that supports the uptake of generic atazanavir/ritonavir.

While the abovementioned arguments and data support generic substitution also for antiretroviral agents, there are still a number of concerns that need to be addressed. First, generic substitution by pharmacists should be accompanied by sufficient patient counselling and providing relevant information to the patient about any change in size, colour, package etc, when a change to a generic is conducted [8]. Furthermore, it is advisable for the pharmacist to inform the other health-care providers of this change. As optimal adherence is crucial for successful HIV therapy, patients should always be informed about changes in drug formulations, and health-care providers must ascertain that patients have understood these changes.

A second issue is that from clinical practice we know that some patients present with (aspecific) side effects shortly after changing formulations containing the same amount of active drug [9]. The mechanisms behind the reported side effects can be divided into at least four subcategories, each of which may require a different intervention. First, it is possible that a generic formulation may contain excipients to which the patient is allergic or experiences complaints. The pharmacist is the appropriate health-care provider to give the patient the correct generic medicine without this excipient. Second, despite proven bioequivalence, there can be subtle differences in rate of absorption between formulations, leading to changes in peak concentrations in selected patients. Third, psychological aspects may play a role, where a patient has lower confidence in the generic product than in the innovator product [10]. Therefore, he may report adverse events, sometimes with the intention to be allowed to resume the innovator product. Fourth and finally, shortly after the switch to a generic drug, other factors in a patient may change too, which can mistakenly be attributed to the generic switch. For example, we had recently a patient who acquired acute hepatitis A infection a few days after initiating generic abacavir/lamivudine. When liver enzymes were increasing and serological evidence of the viral infection was pending, the generic substitution was among the potential explanations, and the physician instructed – understandably – the patient to resume taking the innovator product.

Many Western countries have a health-care system where generic substitution is standard policy to save costs. A pitfall of this system is that patients, physicians and pharmacists are forced to follow this system. The abundance of choice may unfortunately lead to lower acceptance and confidence among physicians and patients [11].

Therefore, we propose that patients reporting adverse events shortly after switching to a generic product, with the exception of a probable or proven allergy to an excipient, should be asked to participate in an N-of-1 study [12]. Pharmacists should be able to provide both the innovator and the generic formulation in a blinded fashion, for example for 1 month, and both the physician and patient are not aware which formulation has been dispensed. At the end of this period both patient and physician can decide on the optimal formulation, and in case of no preference, the generic version will be dispensed. In case of evidence for a preference of the innovator product over the generic (or for one generic over another generic) this should be recorded in the electronic patient record and specified on the physician order. Health-care insurance providers should accept this preference after receiving adequate documentation from the prescribing physician.

In summary, we do not believe that generic substitution of an antiretroviral agent is never a problem. However as outlined above, these problems can and should be addressed in a structured way. Furthermore, adequate counselling of patients by physicians, nurse consultant/specialist and/or pharmacists can diminish problems associated with generic substitution. As a result, we are confident that maximal cost savings will be achieved through generic substitution without harming the patient's treatment outcome.