In this issue of Peer Perspectives in Oncology, renowned oncology investigator Dr. Daniel Von Hoff describes a unique approach to the phase I trial that offers faster drug development. Known as “the complete phase Ib,” the design streamlines phase I by testing various drug combinations in one trial with multiple arms running in parallel. This structure creates rapid accrual rates, substantial economies of scale and significant time savings for sponsors versus the traditional sequential approach. It also garners enthusiasm from investigators and patients, who ultimately gain greater potential for improved therapeutic benefits and care.
(Note: If you’d like to talk directly about a phase Ib for your oncology drug, contact us. We’re a specialty oncology CRO.)
Daniel D. Von Hoff, M.D. is Senior Investigator and Head of Translational Research at the Translational Genomics Research Institute’s (TGen) Translational Drug Development Division and Head, Pancreatic Cancer Research Program in Phoenix, Arizona. He also serves as Chief Scientific Officer for U.S. Oncology and the Scottsdale Clinical Research Institute, and is a founding shareholder of Medelis.
Dr. Von Hoff’s major interest is in the development of new anticancer agents, both in the clinic and in the laboratory. He and his colleagues were involved in the beginning of the development of many of the agents we now use routinely, including mitoxantrone, fludarabine, paclitaxel, docetaxel, gemcitabine, CPT-11, gefitinib and others. At present, he and his colleagues are concentrating on the development of molecularly targeted therapies.
Dr. Von Hoff’s laboratory interests and contributions have been in the area of in vitro drug sensitivity testing to individualize treatment for the patient. He and his laboratory are now concentrating on discovery of new targets in pancreatic cancer. Dr. Von Hoff has published more than 529 papers, 129 book chapters, and more than 891 abstracts.
Dr. Von Hoff was appointed to President Bush’s National Cancer Advisory Board from June 2004 – March 2010. He is the past President of the American Association for Cancer Research, a Fellow of the American College of Physicians, and a member and past board member of the American Society of Clinical Oncology. He is a founder of ILEX Oncology, Inc. (recently acquired by Genzyme). He is founder and Editor Emeritus of Investigational New Drugs – The Journal of New Anticancer Agents as well as the Editor-in-Chief of Molecular Cancer Therapeutics. He is also proud to have been a mentor and teacher for multiple medical students, medical oncology fellows, graduate students, and post-doctoral fellows.
Medelis: Dan, you have evolved an approach to the phase Ib that has repeatedly shown to generate more useful information in less time, getting you to the phase II more quickly. How does it differ from the traditional phase I?
Dr. Von Hoff: The typical phase I approach is essentially serial drug development, which involves running multiple separate trials. Each site at each trial adds a layer of time, cost, and management oversight. What I call “the complete phase Ib” is a simple solution that essentially tests the various drug combinations in one phase Ib with multiple arms run in parallel.
For example, assume you have good preclinical data on a single agent that may be more effective in combination with another agent—for example, a monoclonal antibody plus gemcitabine (GEMZAR®). Most Chief Medical Officers anticipate that the pivotal clinical trials will be standard therapy with or without the new drug, and they prepare for that eventuality by conducting a phase I trial of the combination. Typically, this involves launching five, six, or sometimes more combination phase I trials. Each trial has to be negotiated and managed separately, and each must have its own protocol, adding layer upon layer of cost and effort.
The strategy I’m suggesting eliminates all these separate serial trials because it puts the combinations into one phase Ib trial.
Medelis: One study with multiple arms sounds deceptively simple. Would you provide an example of how this might work?
Dr. Von Hoff: Take the example of a monoclonal antibody. The preclinical information shows great efficacy, and it looks good in separate combinations – with AVASTIN®‚ gemcitabine, DOXIL®, and SUTENT®. Now you’re looking at doing four separate phase I trials to test each combination, with each trial probably costing approximately between five hundred and eight hundred thousand dollars when all is said and done. What we propose with the complete phase Ib involves just one protocol with patients being streamed into different arms depending on what is most appropriate for the individual patient.
Conceptually, here’s how it would work. Mrs. Jones has newly diagnosed advanced disease where the standard treatment is gemcitabine. You’re going to give her gemcitabine anyway, so she gets gemcitabine plus the monoclonal antibody.
Mr. Smith comes in with advanced kidney cancer. He is going to get Sutent anyway, so he’s put on Sutent plus the monoclonal antibody.
Medelis: So at minimum, in the complete phase Ib, every patient is on the standard therapy?
Dr. Von Hoff: Yes, and that alone is especially motivating to patients, since they’re getting the approved drug plus an additional agent that could potentially optimize therapy. Also, most patients who walk into an oncology practice are eligible.
The other advantage is that patients who are early on in their disease immediately enter a phase I trial receiving the best possible treatment. This speeds things up considerably, and it also means the patient population is relatively healthier, having received less previous treatment. This greatly increases the chances of seeing a response in the phase I trial. Investigators, patients, physicians and the sponsor stay more motivated as a result. Of course, there is a chance that you might see more side effects, but we haven’t found that to be the case.
Medelis: What is the dosing strategy in the complete phase Ib?
Dr. Von Hoff: Essentially you’re escalating the dose of your monoclonal antibody or whatever the other new agent may be. For example, I recommend using the full dose of the standard drug as specified on the package insert. Then, on top of the standard doses, we would then use one-third the single-agent dose of the monoclonal antibody in three patients, two-thirds in three patients, and a full dose in three patients.
Medelis: Do you always have three patients at each dose level?
Dr. Von Hoff: Treating three patients per level gives confidence in safety at each level. It only takes three levels at the most because you already know the dose of your investigational agent, and we have the dose for the standard agent.
Medelis: Is it necessary to have animal data on each arm?
Dr. Von Hoff: I like to have some animal model data that establishes the combination effects, ensuring excessive weight loss and other effects do not occur, for instance. If the animals haven’t lost weight, then I’m usually not worried about the combination in patients. As an extra safety measure, we also do abbreviated kinetic sampling.
Medelis: How important is entry criteria in the context of the complete phase Ib?
Dr. Von Hoff: Entry criteria are important and can change the value of an agent under study. If, for example, a monoclonal antibody appears to sensitize to a drug such as Doxil, you might establish an entry criterion specifying the patient can have prior exposure to Doxil. Adding the monoclonal antibody in this context puts enormous value on it, in a sense rescuing the patients.
Medelis: What you mean by “rescuing patients?”
Dr. Von Hoff: A good example involves one of the best trials ever done. It was by Merck AG for the approval of ERBITUX® (Cetuximab). The entry criterion was patients with colon cancer progressing on CPT-11 (irinotecan). Patients were either given Erbitux plus CPT-11, or Erbitux alone.
What was so clever about the trial was that the investigators exploited preclinical evidence that the Erbitux/CPT-11 combination would be rescuing some of these patients, and they established Erbitux as the control.
The trial was highly positive. It showed clinically what the preclinical findings demonstrated — that adding Erbitrux to the CPT-11 dose at which patients’ disease was progressing reverses resistance. There was the additional advantage of greater patient support for this type of phase I since they were getting standard therapy, something an Institutional Review Board is very positive about.
Medelis: How are accrual rates affected with your approach?
Dr. Von Hoff: Compared to the typical phase I approach, the complete phase Ib can produce rapid accrual rates because of patient acceptance. You’re adding an agent to standard therapy and not complicating patients’ involvement by doing things sequentially.
Medelis: Can you discuss the logistical aspects for the complete phase Ib — total number of sites and patients, for instance?
Dr. Von Hoff: With our group, U.S. Oncology Research, we would run a complete phase Ib in up to six sites, depending on the number of arms. Keep in mind, this is not a randomization and each arm could be, say, the monoclonal antibody in combination with any appropriate standard agent. Each arm includes about 9 to 12 patients. If you really know your investigational agent well, you can cut the investigational agent down to half the dose and then the full dose, so it is either one or two escalations. That would mean three patients per level so it could be six patients for evaluation plus an additional 3 to 6 patients to firm up the phase II dose.
Medelis: While the complete phase Ib looks to be more costly than a single phase I trial, are there cost savings compared to conducting sequential phase I trials?
Dr. Von Hoff: The real savings is time. You don’t have to deal with separate amendments. There is only one protocol, creating an operational economy of scale in a sense. In weekly calls, everybody is on the same page. If one arm has a problem — for instance, two out of three people have dose-limiting toxicity — then you substitute in your next arm with an amendment. Previously, if this happened, you would have to shut down that one trial, losing all your start-up costs. It’s a “plug-and-play” modular approach.
The other advantage is that sponsors are happy because the trial gets up and running quickly. The one problem we may have is there are more patients than slots to be filled.
Medelis: In other words, you might be over-accrued?
Dr. Von Hoff: Yes, and it can be a problem because it yields a lot of data, and the CMO has to tease apart the best potential directions to pursue. In this context, a randomized phase II becomes the obvious next step.
Here is a classic example. We have many patients with colon cancer who are on Avastin, an inhibitor of VEGF. Stopping Avastin is problematic because it leaves the VEGF unopposed, potentially leading to a cancer flare. In a case such as this one, clinicians are much more comfortable with a complete phase Ib that has an Avastin arm — if they see activity with the monoclonal antibody plus Avastin, it is extremely important. If the patient was on an Avastin-containing combination for five months before the cancer progressed, and then s/he goes on the monoclonal antibody plus Avastin for up to nine months, what you are in fact doing is changing the natural history of the patient’s tumor. You can see how there is a lot of information to be mined from this approach, giving the CMO many options for phase II.
Medelis: What effect does the complete phase Ib have on the randomized phase II?
Dr. Von Hoff: The complete phase Ib gets you ready for your randomized phase II faster by cutting down time for the whole development program. Instead of a sponsor funding, negotiating, and managing, say, five trial sites, five budgets, five contracts, five protocols, with the complete phase Ib, you test all the possible combinations suggested by the animal systems or cultures in one trial. The first scenario could take a few years; the latter, probably about six months … and that includes doing the trial.
Medelis: Has the complete phase Ib caught on? Is big pharma adopting this approach?
Dr. Von Hoff: In 2007, at ASCO, when we presented the complete phase Ib approach as a way to get to the randomized phase II more quickly, there were only about 25 people talking about it. Now I get a call once a week and we have six trials running. It makes sense for big pharma because it generates the chunk of information you need at one time, rather than serially, for better decision-making.
Medelis: What happens when the complete phase Ib approach pushes for drug combinations that aren’t approved? Who pays for the drug?
Dr. Von Hoff: If you have good evidence for using an agent in an arm but it isn’t indicated yet, then the sponsor would help pay for that drug for that patient. Or there may be literature indicating that an agent works, but the company hasn’t filed an NDA for an extension. In these cases, you at least have a fighting chance with the insurance company. And even if insurance doesn’t pay, the sponsor has a rationale and incentive for covering the cost of the agent.
Medelis: What’s the FDA’s position on this approach? Have they implemented any new regulations to facilitate it?
Dr. Von Hoff: Each trial is filed with an IND and performed with the utmost safety. In reality, it’s not fundamentally any different than doing six trials in serial fashion.
Medelis: Keeping investigators engaged and interested is critical for any trial. How have investigators responded to the complete phase Ib?
Dr. Von Hoff: Investigators are enthusiastic about this approach. One reason is that patient enrollment is extremely fast; now you have the opportunity to treat almost every patient who walks into the clinic. The patient population is relatively healthier because they’ve received less treatment, which increases the chances of seeing a response in the phase Ib trial. Patients, their families and doctors like it because the patient has the greatest chance of getting a positive effect because, in most cases, the drugs are already approved for their type of cancer. And they’re getting an extra kicker — another agent that might help them even more.
Medelis: Looking ahead a few years, what do you see? Will the complete phase Ib become the “typical” approach?
Dr. Von Hoff: I would say that it’s just another drug development methodology. It doesn’t have to become the standard to have utility. I’m hoping there will be even better ideas down the road. Right now, we know it saves time and effort. You see responses. And I think there’s a lot in it for patients, which is one reason it’s catching on so fast.
Medelis: The focus on combinations seems to be a major shift. It wasn’t so long ago that sponsors were pursuing approval for single agents only.
Dr. Von Hoff: Cytostatic agents are okay by themselves but many types are better in combination. Many of the recent approvals are for combinations, potentially increasing the relevance and need for the complete phase Ib. In the old days, you usually got approval as a single agent and then you would start the combination work.
Medelis: How is the focus on combination therapy changing business alliances among companies that are inherently competitive?
Dr. Von Hoff: There are some pioneering companies who are now establishing overarching agreements to facilitate combining their drugs, including investigational drugs not yet approved. It is a powerful edge for two pharmaceutical companies to work together in this manner. It also gives smaller biotechs opportunities to bridge to big pharma.
Medelis: In your experience, how open have companies been to these possibilities?
Dr. Von Hoff: We do not have any trouble with established agents. Most of the time, the sponsor of the standard agent will supply the drug for a study because it could potentially expand the agent’s utility, and they’re getting the rest of the study free. If a biotech company can get the big guys to supply the agent, then that really cuts down on the cost. Everybody benefits.
Medelis: What would happen if you had two investigational agents in an arm?
Dr. Von Hoff: That would take a lot more negotiation. But let’s face it, if two investigational agents look more promising in combination than either agent alone, why would you not want to collaborate? I think those who don’t will be left in a cloud of dust.
Medelis: At what stage in the drug development cycle should a CMO of a small to medium-sized biotech start planning for these potential alliances?
Dr. Von Hoff: Right away, and definitely at the preclinical stage. A CMO should be thinking, what is standard therapy now? What kind of preclinical information will be required to structure one study with multiple arms? This kind of forward thinking can help enormously with funding since you’re talking about saving time and money — you’re funding just one study — and you don’t have to find the needle in the haystack to get patients on these studies. It generates interest in the molecule because you have responses in various combinations. Once you see a response, you have your evidence for moving forward. It sets you right up for the randomized phase II.
Medelis: So the complete phase Ib sounds like a win-win-win since there is benefit to patients, investigators, and sponsors?
Dr. Von Hoff: Approaching the phase Ib as not just a toxicity trial but also a therapeutic opportunity is actually a quadruple win. CMOs benefit on the business side; they get better data faster, helping to drive fundraising and corporate development. Investigators have an opportunity to make an impact, delivering better care and the chance for a better outcome. And sponsors get answers faster and with less bureaucratic hassle since trial management is streamlined.
Last but not least, the complete phase Ib is an important opportunity to truly help patients. And in the end, that’s the most valuable win of all.
Relationships are important to us. Reach out to learn more about how our approach, experience, investigator relationships, bid transparency and flexibility deliver success for our clients.