In January, amidst calls to improve patient safety by optimizing licensed drug formulations, the FDA released a draft guidance that signalled a departure from the most commonly used method of identifying a new therapy’s ideal dosage.
The non-binding draft guidance gives insight into the agency’s current outlook and recommendations for standard and expedited programs in oncology. The agency proposes that pharma move away from the common maximum tolerated dose (MTD) to trial designs that prioritise drug dosage optimisation at different stages.
The focus on identifying the ideal drug dosage is not new. In 2021, the FDA granted Amgen’s (NASDAQ:AMGN) non-small cell lung cancer (NSCLC) drug Lumakras (sotorasib) an expedited approval, making it the first FDA-approved KRAS inhibitor. Since this approval, KRAS inhibitors have risen in popularity with the FDA also approving Mirati Therapeutics’ (NASDAQ:MRTX) inhibitor Krazati (adagrasib) as a therapy for NSCLC. However, Amgen and other pharmaceutical companies have been criticised when they opted to choose the drug’s dosage based on the MTD model, which is commonly used in dose-finding trials.
Most drugs being developed now are more tailored to specific mechanisms of action compared to past drugs, says Dr. Gareth Veal, PhD, a professor of cancer pharmacology at Newcastle University. In the past, a drug may have only generally targeted cell replication. “However, now more therapies target specific abnormalities in tumors,” says Veal. For this reason, many industry experts say the MTD model is lacking in a modern context. “I think there’s absolutely no basis for using the maximum tolerated dose whatsoever,” says Dr. Mark Ratain, director of the Center for Personalized Therapeutics at the University of Chicago. The evidence of its superiority has been rarely proven for any drug, he adds.
While deliberating on Lumakras’s approval, the FDA requested that Amgen test the drug again at a lower dose to understand if the currently marketed formulation is unnecessarily high. The demand came amid concerns that the drug causes a slew of side effects such as nausea, diarrhea, and liver issues. In 2021, Amgen agreed to conduct a Phase II study (NCT04933695), which is supposed to complete in May 2024, according to Clinicaltrials.gov. If the new clinical study finds that a lower drug dose can achieve similar efficacy, the hope is that this could result in fewer side effects for patients. However, this would also lead to a revaluation of Amgen’s financial predictions for the drug, as a lower recommended drug dosage may generate less revenue.
What is drug dose optimisation?
The shift towards the drug dose optimisation model in cancer has been a long time coming. Advocates have pushed for the industry to get on board for the last few decades, because they say the MTD approach opens patients up to high levels of risk from toxicities. In standard and expedited programs, the FDA expects sponsors to adhere to the optimised dose model when appropriate.
In an optimised dose clinical trial, randomised dose-ranging allows companies to find the dose above which there is no evidence of incremental benefit, particularly if higher doses cause increased toxicities. In clinical trials, dose-ranging is testing different drug doses against each other to establish which works best.
According to the National Cancer Institute, the MTD is “the highest dose of a drug or treatment that does not cause unacceptable side effects.” For this approach, a “3+3” model is used where a study enrolls three patients into a given dose cohort. If a subject develops a dose limiting toxicity (DLT) at a specific dose, three more subjects join the same dose cohort, and if more than one in six subjects experience DLTs then the MTD has been exceeded.
The new FDA guidelines say that “the traditional MTD paradigm often does not adequately evaluate other data, such as low-grade symptomatic toxicities (i.e., grade 1-2), dose modifications, drug activity, dose- and exposure-response relationships, and relevant specific populations (defined by age, organ impairment, concomitant medications or concurrent illnesses).” They state that newer drugs like monoclonal antibodies and kinase inhibitors may benefit from a drug optimisation approach due to their different dose escalation profiles, compared to the cytotoxic chemotherapy drugs.
Dr. Serge Cremers, professor of pathology, cell biology, and medicine at Columbia University Irving Medical Center, New York, says the FDA’s preferred approach could be the right move for the industry, but worries about the possibility of underdosing patients “It might improve adherence in the long run, but it has to be a dose that is as effective as the MTD.”
Veal explains that secondary endpoints become more important when optimizing dosage. While planning a trial, the standard Phase I primary endpoint would still focus on safety, but there would be more of an emphasis on showing efficacy in secondary endpoints. He says that, in pharmacokinetic and pharmacodynamic studies, the optimised dose model may be able to predict exposures in the patient population. Thus, Veal also affirms the importance in using this model in paediatric patients to avoid possible long-term side effects.
Veal says adopting this dose optimisation model may mean it takes longer to plan a trial and recruit more patients per dose group. However, in some cases, fewer patients could be needed. For a Phase I study, Veal says, “You may hit your target three or four dose levels below where you’re seeing toxicity, which means that you saved several cohorts from potential toxic effects.”
Impact on the industry
Veal says that for a lot of pharmaceutical companies there is no “big drive to think about optimizing treatment” post-licensing. This often leaves the task to institutions to ensure that licensed drugs are used in their best formulation. And in some cases, studies show that the formulation could be significantly improved. The MTD model made sense for early cancer studies when most drugs had non-specific targets, says Veal.
Ratain says achieving the right dosage requires an understanding of clinical pharmacology and a randomised trial. “One takes knowledge and the other takes time and money.” The clinical pharmacology knowledge also isn’t present at smaller companies, he adds. However, this change has been looming for a long time as the FDA has backed previous dosage optimisation guidance from the International Conference on Harmonization (ICH) E4 guidance.
It’s not the cost that deters many companies, but the time lost from running an extra trial at another dose, says Ratain. “If your patent clock is ticking and it’s going to cost you at least a billion dollars in lost revenue to run that trial, you’re not going to do it unless you have to,” he says. With the guidance, Ratain says the FDA is taking a stand now and “saying ‘sorry, you have to do it.’”
Amgen has disclosed that the Lumakras study has been completed, but has yet to release any results. Ratain highlights that if results show the 240mg dose has equal efficacy to the currently marketed 960mg dose, this may cause Amgen a serious loss in revenue as patients require a lower drug quantity. This may also affect reimbursement agreements with other countries if the pricing changes dramatically. Ratain also poses the question: “If the price of Amgen’s drug goes down by 75%, who will pay full price for Mirati’s drug?” In general however, going forwards, he says companies must write protocols in a “smarter way”.