Portal de Periódicos da CAPES

Orphan drug development is not taking off

2009; Wiley; Volume: 67; Issue: 5 Linguagem: Inglês

10.1111/j.1365-2125.2009.03369.x

1365-2125

Roberta Joppi, Vittorio Bertelè, Silvio Garattini,

Pharmaceutical studies and practices

Although, by definition, rare diseases involve few patients, there are so many – about 7000 (http://www.orpha.net/testor/cgi-bin/OTmain.php?&UserCell=publications) – that their epidemiological impact is impressive: 6–8% of the population is affected worldwide, with 30–40 million patients in the European Union (EU). This makes rare diseases a major public health issue [1]. However, the fragmented market means it is an area of little commercial interest for pharmaceutical companies, and the high price of orphan drugs tends to put them out of reach for the National Health Service [2]. In the EU the European Medicines Agency (EMEA) recognizes the orphan drug status on the basis of epidemiological data (prevalence of the rare disease ≤5/10 000), medical plausibility, and potential benefit [3]. Orphan status designation implies incentives for pharmaceutical companies to develop orphan drugs, including 10 years of market exclusivity in the EU, protocol assistance to optimize development, guidance on preparing the dossier according to regulatory requirements, direct access to and fee reduction for the EMEA centralized procedures, including applications for marketing authorization, inspections, variations, and eligibility for grants from EU and Member State programmes and initiatives supporting research and development. Has the long-term implementation of these facilities been really effective? Since 2000, when the ad hoc legislation [3] came into force, up to 2007, out of 528 designated orphan indications related to 400 orphan medicinal products (OMPs) (http://ec.europa.eu/enterprise/pharmaceuticals/register/index.htm), only 45 (44 drugs) were approved (8.5%) after a mean of 34.8 months (range 6–76 months); out of the 70 OMP marketing authorization applications, 44 were approved (62.9%), 21 withdrawn (30%) and five rejected (7.1%). In the same period, out of 334 applications submitted for 'non-orphan' drugs, 236 (70.7%) were licensed (http://www.emea.europa.eu/htms/human/epar/a.htm). Although preclinical data in the dossiers were fairly satisfactory (Table 1), repeated-dose toxicity studies were not always done in the two recommended animal species [4] for 10 OMPs or with long enough exposure [5] for 24. Lack of genotoxicity, carcinogenicity and reproduction toxicity studies was acceptable for recombinant products mimicking human enzymes (agalsidase alfa, agalsidase beta, aglucosidase alpha, galsulfase, idursulfase, laronidase) or anticancer agents (sorafenib, sunitinib, temsirolimus, trabectedin). In view of the nature of certain compounds, an incomplete toxicological dossier was also justifiable for drugs already on the market for more common indications (e.g. busulfan, ibuprofen and mitotane). Table 2 reports the main characteristics of the clinical studies in the dossier. Out of 44 approved OMPs, 24 (54.5%) received protocol assistance from the EMEA, 16 (36.4%) were authorized under exceptional circumstances, and two had a conditional marketing authorization, which means that the dossier was not complete and the EMEA required additional studies in order to maintain the authorization. Randomized controlled trials were done for 25 products (56.8%). In all the trials but three (5-aminolevulinic acid, deferasirox and porfimer) placebo was the comparator. It was used inappropriately in the case of anagrelide (hydroxyurea), arsenic trioxide (retinoic acid being an adequate control), bosentan, sidenafil and sitaxentan (epoprostenol), cladribine [interferon (IFN)-alpha], imatinib (IFN-alpha), ibuprofen (indomethacin), lenalidomide (bortezomib), miglustat (imiglucerase), pegvisomant (somatostatin), rufinamide (benzodiazepines or newer anti-epileptic drugs such as lamotrigine, topiramate or felbamate alone or as add-on to valproate), zinc acetate (tetrathiomolybdate, penicillamine or trientine), or ziconotide (morphin). In 10 cases the approval was granted with an uncontrolled Phase II study; aglucosidase alpha, anagrelide, dexrazoxane, nitisinone and zinc acetate were authorized on the basis of open-label uncontrolled studies, and carglumic acid on the basis of a retrospective study; for mitotane and betaine only a literature analysis was submitted; this was also the case for hydroxylcarbamide in the paediatric population. In 16 cases the number of study patients was 1000 patients. Whereas for some very rare diseases the small number is justifiable, in other cases it is not: nelarabine was studied in about 100 patients out of 50 000 potential European cases of T-cell acute lymphoblastic leukaemia or T-cell lymphoblastic lymphoma; for Fabry disease the pivotal studies included 41 and 56 patients out of 10 000 potential cases in Europe. Similar figures apply to miglustat, tested on only 28 patients; and clofarabine, tested in 61 patients. Typically the primary end-points are surrogate. Biochemical parameters such as GL-3 (globotriaosylceramide), GB-3 (globotriaosylceramide), IGF-1 (insulin-like growth factor-1), GAG (glycosaminoglycans), homocysteine or ammonia are certainly relevant for the respective diseases, but there is very little proof that the extent of their change is clinically important, justifying long-term treatments. The same holds true for the short-term platelet count reduction by anagrelide. Similarly, the improvement in walking induced by drugs active in pulmonary arterial hypertension, and in mucopolysaccharidosis, although statistically significant, is of questionable clinical importance. The efficacy of anticancer drugs was measured through tumour responses or time to progression rather than survival or quality of life. In some cases the trial was too short in relation to the natural history of the disease: 20 weeks for agalsidase-beta or 18 months for agalsidase-alpha in the treatment of Fabry disease; 12 weeks for pegvisomant acting on resistant acromegaly, for drugs active in pulmonary hypertension or in epilepsy; 4 weeks for anagrelide in essential thrombocythaemia or for sodium oxybate in narcolepsy all seem inadequate. The dossiers of the 20 orphan drugs developed without protocol assistance always contain some deficiencies, but these are also present in 20 out of 24 seeking advice (Table 2). The epidemiological magnitude of rare diseases (7000) is possibly reflected by the number of orphan drug designations (528) but not by their approvals (44), and even less so by their availability on the market (only 26 in the Italian market, driven by a fairly generous national health service). Moreover, clinical and public health needs are poorly met by inadequately documented orphan drugs' efficacy and safety profiles. Limitations include frequent lack of dose-finding studies, often inappropriate clinical design or lack of active comparator where available, insufficient exposure to the treatment, surrogate end-points or weak proof of clinical benefit. The lack of reliable methods for evaluating the effect of drugs on small numbers of patients is also a factor in the general poor quality of the dossiers [6]. However, although less stringent criteria can be considered for orphan drugs than for drugs treating more common diseases, this cannot be an excuse not to guarantee the best possible treatments to patients with rare diseases. For a frequency from 5/10 000 to 5/100 000, at least 25 000–250 000 patients are to be found in the whole of Europe. This should allow adequately sized multicentre trials to test superiority over effective treatments, where available, in terms of clinically meaningful end-points. Longer marketing exclusivity should be granted in order to compensate the long-term follow-up of clinical outcome measures. Because of deficiencies in the documentation, the EMEA required follow-up studies for 18 drugs, which will not necessarily be done and in any case would take many years before the results are available. It is a cause for concern that in spite of an ad hoc law, after 8 years orphan drugs in the EU are still few and poorly studied. Long-term data confirm previous evaluations in the early stages when measures aimed at promoting orphan drugs had still to be fully implemented [7–9], and cast doubt on the effectiveness of the current system. It is not clear what discourages the pharmaceutical industry from developing so few orphan drugs out of the many designated, in spite of the market exclusivity, methodological facilities, and the willingness of the European health systems to pay the high costs and endure the possible low cost-effectiveness of these products. Stakeholders need to reflect on these findings to foster new measures to providw an answer to this largely neglected clinical and social issue. None to declare. This work was funded by institutional funds from the Mario Negri Institute. The authors are grateful to Giuseppina Petruzzelli (G.A. Pfeiffer Memorial Library) and Judith Baggott for editorial assistance.