Introduction

The worldwide economic crisis has put enormous pressure on national health care expenditures, not only in developing countries but also in developed ones. Some have adopted various cost-saving measures to keep pharmaceutical expenses sustainable, which are one of the most variable shares of the health care budget, and thus easier to influence. To back hard choices on pricing and reimbursement, decision-makers increasingly require full economic evaluations (FEEs), which are becoming popular in the general attempt to gain efficiency in the allocation of scarce resources [1].

We take as an example the new drugs for hepatitis C, widely debated at present. Although these drugs seem to be a good step forward in the struggle against this illness, they pose a major threat to pharmaceutical expenditure because of their sky-high prices combined with large target groups. After a brief introduction on the epidemiology of hepatitis C and its treatment options, we critically analyze the FEEs published in the EU on the first generation of direct-acting antivirals (DAAs) indicated for hepatitis C, to assess their contribution to rational decision-making according to the key drivers of their results [2]. We focused on this new subclass of drugs to draw lessons for the forthcoming generation of even more effective anti-HCV therapies, offering both greater efficacy and safety.

Epidemiology and treatment

Hepatitis C is caused by exposure to blood infected with the hepatitis C virus (HCV). The pathology ranges in severity from a mild illness lasting a few weeks to a serious, lifelong condition that can lead to liver cirrhosis or cancer. HCV infection is typically marked by slowly progressive hepatic fibrosis, from stage 0 (no fibrosis) to stage 4 (cirrhosis). Around 75–85 % of the newly infected people develop chronic infection and 60–70 % chronic liver disease, 5–20 % of those leading to cirrhosis and 1–5 % dying from cirrhosis/liver cancer (hepatitis C is the cause of 25 % of liver cancers).

The diagnosis is often lacking because most infected people are asymptomatic and common methods of antibody detection cannot distinguish acute from chronic infection. Thus, not surprisingly, prevalence data on hepatitis C are scarce in Europe and vary a lot from one country to another, ranging from >3 % in Italy to >1 % in France [3, 4].

There are several genotypes of the HCV, which may respond differently to treatment. In Europe, genotype 1 is predominant, followed by genotypes 2 and 3 [5]. The efficacy of therapies is measured by the surrogate endpoint “sustained virological response (SVR)”, used in all the clinical trials (CTs) for approval [6]. SVR is defined as an undetectable HCV viral load 6 months after completing a successful course of HCV treatment.

The current standard of care for chronic HCV infection is a weekly subcutaneous injection of PEGylated interferon α in combination with twice-daily ribavirin (the so-called “double therapy”). Response varies from 66 to 80 % for patients with genotype 2 or 3 to only about 45 % for those with genotype 1 or 4 [7]. Unfortunately, interferon is often not well tolerated because of side effects (anemia, neutropenia, thrombocytopenia, thyroid dysfunction, etc.) so many patients do not complete the course [8].

Two new therapeutic agents, boceprevir (BOC) and telaprevir (TEL), were given market authorization in 2011 by the European Medicine Agency. Both belong to the same subclass of DAAs and are indicated (in combination with peginterferon α and ribavirin) for adult patients with chronic hepatitis C genotype 1 who are naive or have failed previous therapy. The efficacy of these new drugs has appeared to be promising, with an average increase almost double in SVR for genotype 1 patients than with standard dual therapy, although responses are typically lower in those with advanced cirrhosis or other markers of poor outcome [7]. The most challenging complication in both “triple therapies” is still anemia, which occurs in 36–50 % of cases [9].

The regimens of the two triple therapies differ (Fig. 1). Before starting BOC, the patients need to take the double therapy for 4 weeks, after which they add BOC three times a day for up to 44 weeks. TEL should be taken twice a day for 12 weeks, additionally to interferon α and ribavirin. Then, the double therapy will continue for a long period (up to 48 weeks). The total length of the two therapies depends on several factors, such as patients’ previous treatment and results of blood tests during treatment.

Fig. 1
figure 1

Labeled regimens of BOC and TEL [4, 5]

Full size image

In January 2014, EMA approved sofosbuvir, a very promising antiviral which belongs to a new subclass of DAAs [10], and many more products are due to be approved in the short run. For instance, simeprevir or daclatasvir combined with sofosbuvir will be used as an interferon-free treatment. Very recently the CHMP adopted a positive opinion following an accelerated review procedure on a fixed combination of sofosbuvir and ledipasvir in a once-daily single pill. Another triple therapy, interferon-free and fixed-dose, is under accelerated investigation by EMA at present [11].

All these forthcoming therapies should be very effective (with SVR rates over 90 %) and have fewer side effects than the present ones.

Literature review

We did a literature search on the PubMed international database to select FEEs on hepatitis C including the new drugs BOC or TEL, conducted in the EU countries and published in English from January 2011 until March 2014. The search terms used were “boceprevir or telaprevir” and “cost”. We screened the selected articles to assess the main methodological features of the FEEs, using a common checklist based on the one adopted to abstract studies in the EURONHEED database [12].

We retrieved 52 articles: 41 were discarded because they did not include a FEE on BOC/TEL, and were: (a) partial EEs (1); (b) FEEs on other drugs/subgroups or on clinical procedures (5); (c) studies on clinical issues (30); (d) reviews, comments or methodological articles (5). Since five FEEs did not concern the EU setting, we eventually selected six articles [13–18] (Table 1).

Table 1 Main characteristics of studies
Full size table

The analyses, all conducted from a third-party payer’s perspective, concerned only four jurisdictions (Italy, Portugal, Spain, and the UK) since we found two separate articles for both Italy and the UK (one on naive and one on pre-treated patients, written by the same authors). The studies included a CUA and half of them a CEA, too. The studies compared triple therapy with BOC/TEL to standard dual therapy, except in one case where the ‘do nothing’ alternative was arguably adopted, assuming that experienced patients are never re-treated in clinical practice. Half of the studies included a direct comparison between the two triple therapies as well, although head-to-head CTs are lacking. All studies but one were (directly or indirectly) sponsored by the pharmaceutical industry and three were co-authored by one employee at least.

From a methodological point of view, all studies were based on modeling with a long-term horizon, using virtual cohorts based on CT patients (in two studies even a purely hypothetical cohort), which hardly reflect the real epidemiology of the settings analyzed. Although efficacy was mainly derived from CTs used for registration, all but one used treatment algorithms different from the labels. The probability of disease progression varied a lot from one study to another. Four relied on foreign scores and expert panels to estimate utility. In all studies we found very weak sources like assumptions and expert panels to estimate resource consumption; one Italian study even used the US prices of BOC and TEL as unit costs. Costs of side effects were lacking in two studies.

All studies concluded in favor of triple therapies. Sensitivity analyses highlighted drug prices, discount rate, efficacy, and utility as the most influential variables on baseline results.

Policy implications

Although BOC and TEL have been shown to be more effective than standard therapy, the actual usefulness of the FEEs evaluated, mainly populated by weak sources and based on long-term modeling designed for subgroups of patients and arguable therapeutic regimens, remains uncertain from the health authorities’ viewpoint, as underlined by NICE in its separate assessments of BOC and TEL [19, 20].

Assuming that the forthcoming drugs are even more effective and much safer (needing shorter regimens not necessarily with interferon), we wonder whether future FEEs on these therapies focused on specific subgroups of patients really add any value for decision-making, particularly in a value-based pricing perspective. Despite the sky-high prices of all these incumbent drugs, we are sure that companies would manage to select the most suitable inputs to feed long-term models and pick subgroups with different regimens to show an acceptable incremental cost-effectiveness ratio for each of them. However, when treating larger target groups, as seems the most likely scenario in hepatitis C, we believe the real “crux” of the matter will be the difficulty in sustaining the budget impact of these combination therapies, even in developed countries with different prevalences of hepatitis C. For instance, although the financial burden should be much lower in France than in Italy (according to the hepatitis C prevalence), the French government has announced a plan to selectively tax pharmaceutical companies should their total revenue on hepatitis C drugs exceed a fixed yearly amount [21], while in Italy AIFA (the Italian agency for drugs) has just signed a confidential deal with the sofosbuvir marketer after months of hard negotiations, without issuing any information on the agreed price so far.

In general, we hope national authorities will play a double role in this field. Before deciding on pricing and reimbursement of the forthcoming drugs, the main decision should be what subgroups to treat, trying to separate patients who can wait until the arrival of better treatments after early detection from those who should start treatment as early as possible to prevent severe complications in the future.

To treat as many patients as possible, national authorities could offer industry a sort of “block contract”, a budget estimated from the health care costs potentially avoided thanks to these new therapies. Setting an “average price per volume” (i.e., the ratio between the total budget and the number of patients to treat as a target) should reflect the trade-off between increasing pharmaceutical costs and future inpatient savings for health care services. Like what happened at the end of the last century in a different epidemiological context with the DAAs for HIV/AIDS [22], these new combination therapies, which come from the same drug pipeline, could prevent many future hospital admissions by eradicating the HCVs. The main clinical challenge in such an exercise would be to set the “volume” of treatable patients, taking into account that hepatitis C can be a silent pathology for years, but that liver damage cannot be repaired. This volume could be stratified according to disease severity, hence the urgency for these therapies.

From the supply side, because many companies have stepped into this area, so many drugs should soon be available, broad negotiation could lead industry to accept the “sustainable” prices for the combination therapies, set as above. Alternatively, failing to achieve a “gentleman’s agreement” with the different companies offering combination/single marketed therapies, national authorities could opt for tendering as a valuable alternative. A sufficient number of regional tenders, like what might happen in Italy, would prevent the threat of generating dominant positions and thus the risk of market failure induced by centralized public procurement [23]. A major hurdle could be to assume all these combination therapies of equivalent efficacy. However, real clinical differences could be addressed by adding a ‘quality score’ in tender clauses for specific groups of patients, with the aim of maximizing cost-effectiveness rather than simply minimizing cost. It is also worth noting that these new hepatitis C drugs are not manufactured through complicated processes, an argument often put forward by industry (e.g., for human antibodies) to justify high prices. Their prices can be affected more by trade agreements when the company that discovered and then developed the drug is not the one that markets it, so the marketer has to pay high royalties to the manufacturer.

To conclude, we think new technical solutions are now needed for pricing therapies that are as innovative as they are expensive, to make their total cost sustainable in a period of unprecedented financial crisis for health care services. Here, starting from the promising combination therapies against hepatitis C which are to be launched in the near future, we put forward a very general proposal open to debate, based on real financial budgeting rather than economic evaluation.