By Rabiya S. Tuma, PhD

New drugs begin as good ideas and scientific observations, but the man-hours, people, and perseverance required to turn a compound into a marketable drug extend far beyond that initial effort.

In the case of Bexxar® (tositumomab), a radioactive antibody-based therapy for non-Hodgkin’s lymphoma (NHL), the process took almost 12 years from the time the product first entered human clinical trials until the drug gained U.S. Food and Drug Administration (FDA) approval.

A long time? Too long? That depends on whom you ask.

But Bexxar’s story provides a useful window into the complexities of drug development, which, in this case, began 25 years ago with the creation of monoclonal antibodies.

By the late 1970s, researchers had developed the technology to create monoclonal antibodies, genetically engineered proteins that mimic the body’s own infection-fighting proteins produced by white blood cells. Because the man-made antibodies could be produced in great quantities in laboratories, it made them a viable therapy for large numbers of patients, including those with NHL, a broad term for a collection of cancers that affect lymphocytes (also called B and T cells), some of the body’s main infection fighters.

When patients develop a B-cell type NHL, their B cells multiply abnormally. Used alone or with an attached toxin or radioactive isotope, monoclonal antibodies that recognize B-cell specific proteins lock on to the specific protein (antigen) on the surface of B cells, destroying the circulating lymphocytes, including the cancerous cells. Those involved in the early research were giddy with the promise that these new engineered antibodies would become a cure for all kinds of disease, including cancer.

Lee Nadler, MD, now senior vice president of Experimental Medicine at the Dana-Farber Cancer Institute in Boston, Massachusetts, was fresh from his medical training when he began working with monoclonal antibodies to learn how lymphomas arise.

He teamed up with former classmate Phil Stashenko, DDS, who already knew the monoclonal antibody technology, and together they generated a large number of antibodies, including one they called B1, which has become the backbone of Bexxar.

B1 stood out amongst the many antibodies the team made, says Dr. Nadler, because it binds to a protein found on the surface of all mature human B cells, including most lymphoma B cells, but not to the stem cells that are required for generation of normal B cells.

“We had this antibody that was clearly B-cell specific,” recalls Dr. Nadler, “but we didn’t understand the function of the B1 protein. And because we didn’t know, to be quite honest, we did two things at that point: We tried to figure out what it did, and we tried to figure out if we could use the antibody for treatment.”

But, says Dr. Nadler, after being told by several biotechnology companies that there was no significant market for monoclonal antibodies against B-cell lymphoma, they sold the rights to a company called Coulter Corporation.

Initially, the B1 antibody was used primarily for diagnostic purposes. In 1989, they initiated the FDA licensing procedure by filing an investigational new drug application (IND). The license allowed B1 to be tested in humans. B1 antibody was initially used to clear lymphoma cells from bone marrow harvests prior to bone marrow transplants.

Meanwhile, Dr. Nadler and other scientists were testing additional antibody therapies to turn them into more potent therapies against cancer. One approach was to attach a radioactive molecule or toxin to the antibody. With such a strategy, the scientists could rely on the antibody to travel through a patient’s system until it found its antigen, which, in the case of B1, was B cells. It could then deliver the toxin directly to the cell.

A Drug Is Born
Two groups took up the challenge in the early ’90s of turning the B1 antibody into a functional drug. Oliver Press, MD, PhD, at Fred Hutchinson Cancer Research Center in Seattle worked with very high doses of radiation delivered by radioactive iodine that was attached to B1 followed by bone marrow transplant. Mark Kaminski, MD, and his team at the University of Michigan at Ann Arbor reasoned that if they used a well-titrated dose of radioactive iodine-labeled B1 antibody, they would be able to selectively target and kill the cancer cells while doing minimal damage to the rest of the blood and lymph cells and thus prevent the need for a bone marrow transplant.

“We weren’t looking for remissions,” says Dr. Press about his protocol. “We wanted a cure.” Thus far, the team has treated about 140 lymphoma patients with high doses of radioactive iodine-labeled B1 with a success rate of 40-50%. Many of these patients have never had disease recurrence, including some who had the treatment 10 years ago. This technique, using high doses of radioactive B1 followed by bone marrow transplant, requires a highly specialized team of physicians, nuclear pharmacists, and nurses, and is currently being done in Seattle and a few other centers.

Dr. Kaminski’s approach using lower doses of radiolabeled B1 did not require bone marrow transplantation and can be performed at many regular hospitals and cancer centers.

“When we started our clinical trials,” says Dr. Kaminski, “we noticed really within the first several patients that we were getting results that we hadn’t seen with our prior work with radiolabeled and unlabeled antibodies. We were starting to see complete remissions, even in patients with bulky disease where we wouldn’t have expected to see such responses with other compounds.”
In fact, results from Dr. Kaminski’s phase I trial were remarkable, with six out of nine patients with B-cell lymphoma achieving a partial or complete response to the new therapy.

“It was beyond our expectations—and the expectation of others,” says Dr. Kaminski. “The other thing that was notable about the treatment was that there was very little toxicity, even myelotoxicity,” which meant the patients were going into remission without needing bone marrow transplants.

Both Drs. Press and Kaminski published their preliminary results in The New England Journal of Medicine in 1993, but it was Dr. Kaminski’s approach that subsequently attracted the most attention.

For the next several years, Dr. Kaminski’s team continued to work out the best delivery method, determining that the safest way to deliver this drug would be to tailor an individual dose for each patient by doing a small pre-dosing test instead of simply giving a dose based on a patient’s weight. This individually tailored approach meant the Michigan team required several years to fine-tune the system before they were ready to initiate multicenter trials. During this time, he says, they had quite a number of discussions with the FDA on how to devise a practical and safe dosing scheme.

Expanding the Trials
During the early studies it became clear, says Dr. Kaminski, that more than half of the patients with low-grade or follicular lymphoma responded to the treatment and about 20% of patients had complete disappearance of their disease lasting several years, which was a remarkable improvement over standard chemotherapy. It was time to move forward.

In 1995, Coulter agreed to fund the trials, but because they didn’t have the infrastructure in place to either fund or supervise clinical trials, the company raised venture capital and started Coulter Pharmaceutical, Inc. to oversee the drug development process of tositumomab, the generic name given to the B1 antibody. Tositumomab labeled with radioactive iodine was given the trade name Bexxar.

After small multicenter studies referred to as pilot studies, the team found that other centers could reproduce Michigan’s successful results. Now the time came to devise a strategy that would convince the FDA that Bexxar was better than existing treatments.

This was a bit of a conundrum, says Dr. Kaminski, because an investigational drug is compared to the current standard of care in a head-to-head randomized, controlled trial to determine whether the new treatment improves on the old one. But with the flexible standard of care for lymphoma patients at the time, it wasn’t clear what Bexxar should be compared to.

Given this situation, Dr. Kaminski, Coulter Pharmaceutical, and the FDA came up with a somewhat unusual clinical trial design that measured each patient’s response to Bexxar in comparison to the previous chemotherapy regimen.
Sixty patients enrolled in the pivotal study and were treated between November 1996 and March 1998. Thirty-nine (65%) had either a partial or complete response to Bexxar, which was significantly more than the 17 patients (28%) that responded to their previous chemotherapy regimen. Also, the average duration of response to Bexxar at 6.5 months was significantly longer than had been seen with the previous chemotherapy regimens at 3.4 months. Of the 12 patients (20%) who had a complete response to Bexxar, nine were still disease-free three to four years later.

Regulatory Struggles
In December 1998, Coulter Pharmaceutical signed an agreement with SmithKline Beecham (which merged with GlaxoWellcome in 2000 to become GlaxoSmithKline) to jointly commercialize Bexxar.

By June 1999, 10 years after they filed the initial IND, Coulter Pharmaceutical felt they had enough data to file a biological license application (BLA) with the FDA that would allow the company to market the drug for treatment of low-grade lymphoma. In August, Coulter learned that the FDA refused to file the BLA and requested more information from the company.

In November of 2000, the FDA accepted the second BLA, only months before Corixa Corporation, a Seattle-based biotechnology company run by a team of experienced drug developers, bought Coulter Pharmaceutical and took over the FDA process.

In March of 2001 the FDA completed its initial review of Bexxar’s BLA that Coulter had submitted the previous September and found the data insufficient to warrant approval. The FDA wanted to see additional data from recent clinical trials comparing Bexxar to Rituxan® (rituximab), a non-radioactive antibody therapy that had recently been approved for NHL that targets the same B-cell protein as Bexxar.

“[The FDA was] also looking for extended safety data beyond what was [originally] called for in the clinical trials protocol that the agency had previously approved,” says Steven Gillis, PhD, chairman and chief executive officer of Corixa.

Dr. Gillis suggests one reason for this was that Rituxan had been approved since the Bexxar trial had been designed and had changed the standard of care for NHL.

For the next six months, Corixa’s team collected and assembled data requested by the FDA. Because some of the safety data needed was long-term follow-up information about the effects of radiation on patients’ thyroid function, it meant the company had to track down past patients, run new lab tests, and analyze the results. The agency also requested more information regarding the procedures used in the manufacturing process of the drug, which during its history had been produced in three different facilities.

In September 2001, six months after the FDA’s review letter, Corixa filed their complete response. By October, the FDA agreed that all of the concerns discussed in their review letter had been addressed and they would now reconsider the drug application. Corixa now had six months to wait.

“There was a fair amount of dialogue going back and forth during this period,” says Dr. Gillis. “However, we received a letter on March 12, 2002 that was not a good letter to get.” The FDA concluded again that the drug was not approvable.

This second letter of rejection from the FDA left Corixa with few options, because the agency was requiring more clinical trials, which would take years that the company didn’t think they had, says Dr. Gillis. This was particularly true because the FDA had now approved not only Rituxan but also the radioactive antibody therapy Zevalin™ (ibritumomab), a drug some see as quite similar to Bexxar.

At this point, the company asked for another meeting with the FDA, and after much negotiation and an appeal, the agency agreed to put the drug back into an active review status.

“We made a persuasive argument that Bexxar provided an unprecedented durable response and I think that got some people’s attention at the FDA meeting,” says Dr. Gillis. With the review process reopened, Corixa now had the opportunity to present their data to the Oncologic Drugs Advisory Committee (ODAC), a committee comprising non-agency specialists who review clinical and safety data and advise the FDA on whether or not to approve a drug (see sidebar).

The ODAC panel met Dec. 17, 2002. Corixa and the group of clinical trial specialists presented the results from numerous trials, including data on Bexxar’s efficacy in Rituxan-refractory patients.

At the end of the day the panel voted 10-3 in favor of approving the drug for low-grade lymphoma patients who had failed Rituxan therapy and voted 13-0 in favor of approving it for patients who no longer respond to chemotherapy. “So we got a wildly enthusiastic endorsement of everything we wanted,” says Dr. Gillis.

However, the ODAC panel is only an advisory panel, so the wait for Corixa, Drs. Kaminski and Press, and for patients continued. The agency’s final decision was due in early May 2003.

Approval
On May 5, just as the Bexxar team expected the final approval, they received news of one more delay: The FDA wanted to see the most up-to-date safety information, data which had been alluded to at the ODAC meeting and provided by Corixa, but which had not been fully analyzed by the agency at the time. And the agency said they would need three more months to evaluate the information.

Finally, on June 30, 2003, Dr. Kaminski walked into his office to find a big “Congratulations” sign on his door. The FDA had approved Bexxar.

“Still, after hearing the news, after so many years, I had to see it in black and white before I really believed it,” says Dr. Kaminski. One of the first calls he made was to Pat Haut, one of the first patients treated with the drug more than 10 years ago .

As for what happens now, Drs. Gillis and Kaminski say they are pleased the drug is now available to follicular NHL patients nationwide, but the work hasn’t stopped.

“Now the key is to discover where to place treatment in our overall strategy,” says Dr. Kaminski. “We need to find out if it is best used only when a patient becomes refractory to Rituxan treatment or if we should use it earlier in the disease when we see greater success rates. But those success rates have to be balanced with safety concerns.”

So while the drug is now available to patients who fail Rituxan therapy and chemotherapy, researchers will continue to design and enroll patients in clinical trials to determine how best to use Bexxar.