Nature Biotechnology 26, 1057 - 1059 (2008)
doi:10.1038/nbt1008-1057
Ark floats gene therapy's boat, for now
Randy Osborne1
1. Mill Valley, California
2. In August, gene therapy's turbulent ride through the clinical rapids took a new twist as Ark Therapeutics released positive top-line results from a phase 3 trial of its adenoviral gene therapy Cerepro (sitimagene ceradenovec) for malignant brain tumors. Although the news boosted the London-based firm's shares, the course to market authorization and registration remains strewn with uncertainty—as Introgen, of Austin, Texas, found, to its cost, when the US Food and Drug Administration (FDA) recently refused its Biologics License Application (BLA) for Advexin (contusugene ladenovec), an orphan-designated adenoviral gene therapy for treating head and neck cancer and Li-Fraumeni syndrome.
3. Ark's 'study 904', which was approved by the UK Gene Therapy Advisory Committee in 2004, randomized 236 people with brain cancer to receive Cerepro plus standard care or standard care alone, which consists either of surgery and radiotherapy or of surgery and radiotherapy plus the alkylating drug Temodar/Temodal (temozolomide) from Schering-Plough in Kenilworth, New Jersey. Subjects given Cerepro and temozolomide showed a 42-day improvement over standard care in median survival, reaching significance (P < 0.032). Side effects hemiparesis, aphasia and pyrexia could be blamed on ganciclovir, which is part of the Cerepro protocol and is "pretty toxic," says analyst Stephen Dunn of Boca Raton, Florida–based securities firm Dawson James.
4. The data, due for a full airing at the European Association of Neuro-Oncology in Barcelona as Nature Biotechnology went to press, seem strong. Cerepro—which consists of the herpes simplex virus gene for thymidine kinase (TK) encased in an adenoviral vector in which the E1 and part of the E3 regions have been deleted to prevent replication—was tested in people with operable, high-grade malignant glioma. Doctors injected Cerepro into the cavity left by the removed tumor during the surgery. In the following days, physicians administered ganciclovir—Basel, Switzerland-based Roche's approved drug for cytomegalovirus (sold under the brand names Cytovene and Cymevene). The transformed cells expressing TK convert the prodrug ganciclovir into highly toxic deoxyguanosine triphosphate, which kills any remaining cancer cells. "All the TK does is provide a target for a second drug," says Dunn. "It's an interesting way of doing it, and safe."
5. A more detailed verdict on Cerepro is expected in January of next year, when data on mortality, one of study 904's secondary endpoints, will be reported; 45% of study participants remained alive at the start of August. It seems likely that Ark will use the trial results showing significant improvements in median survival compared with various control groups to apply for a new marketing authorization for Cerepro in glioma in Europe. In 2006, the European Medicines Agency's (EMEA) European Committee for Medicinal Products for Human Use returned the company's previous marketing application, which had been based on a small phase 2 trial. But, as EMEA's deputy head of sector for safety and efficacy Marisa Papaluca Amati is quick to point out, "It was a withdrawal, not a rejection."
6. In the meantime, another frontrunner in adenoviral gene therapy, Introgen's Advexin, has hit a snag at the FDA.
7. Paradoxically, while the EMEA accepted Introgen's marketing application for Advexin (a recombinant, E1-deleted serotype 5 adenoviral vector encoding the p53 tumor suppressor), about a month later the FDA said that the company's biologics license application (BLA) was incomplete.
8. Safety probably is not the issue in the FDA's refusal to accept Advexin's BLA, Dunn says. It could be the prospective biomarker analysis they used in the trial. "I'm wondering if the FDA didn't just go back on their word," after claiming that such data would be acceptable, Dunn ponders. This is significant because Introgen specifically designed their phase 3 trial to prospectively segment patients according to p53 abnormalities and p53 protein levels in pretreatment tumor samples (the company declined to reveal the identity of the mutations).
9. Preliminary results from this open-label, multicenter, randomized study, which compared Advexin with the standard-of-care methotrexate in 123 people with end-stage head and neck cancer, were released in May at the American Society of Gene Therapy meeting in Boston. In the intent-to-treat population, median survival was 6.1 and 4.4 months in the Advexin and methotrexate arms, respectively, which was not significant. Even so, biomarker analysis revealed that survival was increased in Advexin-treated subjects with favorable p53 profiles compared with those with unfavorable p53 profiles (7.2 versus 2.7 months). Introgen declined to comment on the FDA's decision to turn down the BLA and indicated that talks to remedy the situation are ongoing.
10. According to Antonio Giordano, a pioneer in gene therapy and professor of molecular biology at Temple University in Philadelphia, who also uses adenoviral vectors in his research, the FDA's thumbs-down was "not a surprise." The viral gene therapy approach in general "presents lots of well known limitations," he says, especially with regard to immune responses.
11. Indeed, previous instances of wayward inflammatory responses in people receiving viral gene therapies have been at the root of some of the field's darkest moments. In 1999, Jesse Gelsinger died of massive organ failure after receiving a high dose in his hepatic artery of an adenoviral gene therapy for ornithine transcarbamylase in a trial at the University of Pennsylvania (Nat. Biotechnol. 23, 519–521; 2005). Thus far, such adverse events have not been a concern for either Ark's or Introgen's adenoviral therapies.
12. But immune responses have compromised some adeno-associated virus (AAV) gene therapies. In one case, media coverage implicated Targeted Genetics of Seattle, Washington's AAV-2 gene therapy for rheumatoid arthritis in compromising the immune response of a participant in a phase 2 trial; however, the US National Institute of Health's Recombinant DNA Advisory Committee has since clarified that Jolee Mohr's death by histoplasmosis was much more likely due to the tumor necrosis factor (TNF)- inhibitor she was taking systemically rather than the locally delivered AAV vector carrying the gene for the TNF- receptor. T cell–mediated destruction of AAV-transduced cells has, however, been thought to account for self-limited and asymptomatic liver toxicity reported in two subjects on a human factor IX gene therapy for hemophilia (Nat. Med. 12, 342–347, 2006) and elevated levels of creatine phosphokinase seen in certain people receiving Glybera (alipogene tiparvovec) from Netherlands-based Amsterdam Molecular Therapeutics, an AAV-1 gene therapy to treat lipoprotein lipase deficiency. Amsterdam has since gone on to complete enrollment and treat its last patient in the phase 3 study of its orphan-designated gene therapy—and an application for marketing authorization is expected later this year.