SAN DIEGO – Although "it's going to be a bit of a bumpy road" and the process "won't be as pretty as we would like it to be," clinicians have reached "a point where I think we can start to mix and match the appropriate mechanisms" of immuno-oncology drugs to fit individual patient needs, said Tim Reilly, head of early asset development in oncology for Bristol-Myers Squibb Co. (BMS), of New York.

Reilly spoke to an overflow crowd of about 300 attending a panel on the prospects of using I-O drugs as first-line therapy. He compared the progress of still-new therapies to plumbing. "The first time you buy a house and your sink breaks, you have to go and get the tools that are going to be used to fix that sink, you realize that, 'Oh, I have to go back to Home Depot or Lowe's again and again and again,'" he said, "It's a learning, iterative process."

Helping the personalized push is Atreca Inc., of Redwood City, Calif., which has found a way to measure and analyze the structure of clinically relevant immune responses, going on to identify the antibodies, T-cell receptors, and targets that can bring successful treatment outcomes. The company's Immune Repertoire Capture technology profiles a patient's immune response at the single-cell level by high-throughput means to provide "a complete picture of the ongoing immune response across time" as the patient undergoes treatment, said Atreca's Chief Scientific Officer Norman Greenberg. "Why would you rob a bank? Because that's where the money is. We're looking at patients who survived their cancers as a source of antibodies," he said.

Atreca has processed more than 150 "elite or exceptional responders," ransacking them for "antibodies and/or T cells that are responsible for their remarkable survival," Greenberg said. "It's the ultimate in personalized medicine – a patient's own antibodies, but now we're taking them out of the context, turning them into drugs, and we can put them into vials and treat multiple patients at pharmacologic doses."

The company has oncology and infectious-disease programs at the discovery stage. "Some of you are thinking, 'Great, we're probably identifying a lot of antibodies, neoepitopes, things that arise spontaneously and may not be generally expressed in the population,'" he said. "Let me tell you that a lot of the antibodies we're finding are in fact against public antigens." Patients are showing changes in B-cell and T-cell repertoires as they are given courses of I-O therapies.

Meanwhile, a "huge number of clinical trials" in the space are moving ahead, noted Kaan Certel, head of external innovation in oncology with Paris-based Sanofi SA. "We really need a rationale around how we're going to combine things, and currently, in the excitement of everybody wanting to be the first one to come up with the best combination, there really is no rationale out there," he said.

"Today, immunologists are saying, 'I told you so' to oncologists. Oncologists would say, 'Well, there have been targeted therapies and there have been chemotherapies that have worked in people before and in fact cured some people. What about strategies combining those targeted therapies [and chemotherapies] with I-O?'" But a grasp of the science is still lacking. "We now know that we don't understand the molecular basis of many of the cancers," he said. "We know that they are very diverse." Genome sequencing and data mining will solve some of the problems.

'Big, risky bet' in cancer vaccines

Another recent leg up is provided by reverse translation, said Kinney Horn, director of business development for Genentech, part of Basel, Switzerland-based Roche Holding AG. "We're actually seeing information coming out of the clinical trials, out of the patient experience, that is informing the research," he said. "The [animal] models are going to be there and, frankly, they're getting better and better. But what is different, maybe, over the last five years is that we're seeing information coming back that we can analyze from the human patients themselves. That is directing our research efforts. We have not quite seen anything like this before."

Moderating the panel was Tauseef Butt, president and CEO of Malvern, Pa.-based Progenra Inc., doing early stage research into the ubiquitin-proteasomal pathway and cellular protein regulation. The company has turned up candidate inhibitors of the ubiquitin pathway for potential use in cancer, inflammatory disease and other therapeutic areas.

An audience member asked how a startup company might be expected to put together the "incredibly expensive preclinical package" necessary to secure a deal with one of the major players working with I-O combos. Butt said "the science is really not very good, sorry, at small companies. 'Knock my socks off, give me the experiment.' I think that is what's being said" by panelists. But Sanofi's Certel said that "even failed trials are useful information."

Another audience member asked how exquisitely the mechanism of action must be known of a would-be drug, given that some are so complex. "I would say it's critical, to be honest with you," said BMS' Reilly. "Something that looks like it works but you don't actually understand how it works is far less useful" than more precise knowledge. "If you don't understand that, in my opinion, that's taking us back 20 years. We want to go forward."

Genentech's Horn pointed out that "drugs that are doing something to tumors are probably doing things to healthy cells, too." Having "mechanistic data and a bit of safety data" becomes particularly important in work with combos, he said, adding that his firm is often inspired to make deals based on "scientists getting excited about the mechanism" as they know it.

Asked about new modes in pipelines, Horn said his firm "made a very big bet, it's a risky bet" in Biontech AG, of Mainz, Germany, in September 2015, paying $310 million in up-front and near-term milestone payments. The deal is geared for developing cancer vaccines.

"We think that, if this does work, it would be the beginning of patients having their tumors sequenced, characterized at an individual level, and then a vaccine prepared just for them very quickly to be combined with a lot of the other marketed agents for a truly personalized approach to cancer," he said.

"We follow the science in everything we do, but I think going back to the individual patient is probably the path to above 30 percent [efficacy of I-O drugs among patient populations] to sustained cures."