The conditional marketing authorization granted to AstraZeneca PLC’s cancer drug Tagrisso meant significant challenges for the health technology assessment bodies evaluating the product, according to Chris Hoyle, the company’s director of health economics & payer analytics (oncology). He says AstraZeneca found a yawning gap between requirements for early regulatory approval and positive reimbursement decisions because HTA agencies took different approaches to uncertainty, indirect comparisons and immature data.
“Accelerated products are less likely to fulfil the requirements of a rigid algorithmic framework for assessment” – Chris Hoyle, AstraZeneca
In February 2016, Tagrisso (osimertinib) won EU conditional marketing approval for treating locally advanced or metastatic epidermal growth factor receptor T790M mutation-positive non-small cell lung cancer, following an accelerated assessment procedure. The conditional approval was based on two single-arm Phase II trials. Full approval came in April 2017, supported by a confirmatory Phase III randomized control trial (RCT) that compared Tagrisso with platinum-based doublet chemotherapy.
However, prior to full approval, the accelerated assessment and conditional marketing authorization posed significant challenges for HTA agencies assessing the drug, explained Hoyle. For example, the evidence HTA bodies had to review was based on a single-arm trial, which yielded no comparative data. In addition, agencies struggled with the immature overall survival data, he said at the ISPOR 20th Annual Congress in Glasgow, UK, on Nov. 8.
The company therefore had to try to find solutions to “bridge the gap between accelerated approval and the time when there is sufficiently matured evidence to fulfil the requirements of HTA agencies to make a final decision,” said Hoyle. To support its HTA submissions, AstraZeneca needed to provide comparative evidence of the benefits of Tagrisso over chemotherapy. It used data from a study of patients treated with chemotherapy and put together an indirect comparison of the two treatments.
But not only did the HTA agencies have difficulty in dealing with the evidence following conditional marketing authorization, they also had different ways of approaching it.
HTAs At Odds Over Tagrisso
England’s NICE, for example, published its first take on the drug in May 2016. It declined to recommend Tagrisso, claiming that overall survival had not been adequately demonstrated. However, the company subsequently provided an updated analysis including results of the RCT. Thanks to that new analysis, in October that year NICE was able to publish guidance recommending that the drug be made available through the Cancer Drugs Fund. Hoyle pointed out that NICE accepted the indirect comparison as valid.
Meanwhile, also in 2016, Germany’s authority in charge of pricing and reimbursement, the G-BA, initially concluded that Tagrisso offered no additional benefit over chemotherapy. It acknowledged that the indirect comparison had been well conducted but would not accept it or the results from the single arm trial. Instead the G-BA said it required comparative data from a randomized control trial.
“The implications of this decision to us were that a single-arm trial and innovative methods of analysis were not valid. There was also a profound difference of opinion on clinical benefit between the G-BA and the regulator,” Hoyle said. However, in 2017 AstraZeneca resubmitted with evidence from the RCT and the G-BA awarded the drug a rating of significant additional value. According to Hoyle, quality of life and safety data also featured in the decision making.
In France, the HTA agency HAS concluded that Tagrisso did not offer any added benefit, although when RCT data became available, the rating was improved to minor added benefit. “HAS did not accept the indirect methodology, they only accepted the data from the RCT,” said Hoyle. Furthermore, the agency did not deem quality of life data to be valid.
Closing The Gap
To close the gap between requirements for regulatory and reimbursement approvals, Hoyle wants to see better alignment between HTA bodies and regulators regarding evidence needs. “HTA agencies should specifically accommodate priority medicine designations and conditional marketing approvals,” he said.
Moreover, to reap the benefits of accelerated development, payer and HTA agencies must recognize the value of different types of data beyond RCTs. They must also consider alternative methods of evidence synthesis and of balancing the management of uncertainty with the magnitude of the likely treatment effect, according to Hoyle. “Accelerated products are less likely to fulfil the requirements of a rigid algorithmic framework for assessment.”
Furthermore, he said that companies, payers and regulators should engage with each other earlier to look at how to bring flexibility to the process and stagger the evidence. Early dialogue, ongoing scientific advice and an EU relative effectiveness tool could go a long way to overcoming some of the challenges, he commented.