When Drug Trials Are Halted

After three patients with B-cell acute lymphoblastic leukemia suffered fatal cerebral edemas during Juno Therapeutics' phase II trial of a CAR-T therapy in July, company officials moved quickly to stop the trial and alert the FDA. Regulators responded by issuing a full clinical hold on the trial—halting treatment of enrolled patients and preventing new accrual—pending an investigation by the company.

Clinical holds have the potential to significantly delay or even end an experimental drug's path to market, but Juno's experience turned into a best-case scenario. Regulators lifted the hold 3 days later, after Juno officials supplied data indicating that the deaths were due to an overly intensive chemotherapy regimen of fludarabine and cyclophosphamide given prior to the CAR-T therapy. The trial resumed after fludarabine was dropped.

Although single-agent cyclophosphamide had been used to debulk tumors in earlier trials, some evidence suggested that adding a second agent could lead to better expansion and longer persistence of CAR-T cells in peripheral blood, says Mark Gilbert, MD, chief medical officer for Seattle, WA–based Juno. “All three deaths occurred soon after patients started on this regimen, and a thorough investigation indicated that the conditioning intensity with fludarabine and cyclophosphamide was the culprit.”

“The critical piece in resolving this quickly was having access to datasets from earlier trials at our partner academic institutions that we could use for comparative analysis to help us isolate the problem,” says Gilbert. “At the same time, we were motivated to act quickly by the extraordinary medical need of the patients and our strong belief that this therapy had potential benefit, despite its risks.”

There is no average duration for clinical holds, say FDA officials, although many last longer than a few days and have the potential to permanently derail a drug's development. While acknowledging the frustrations that holds can cause, however, they note that safety is the overriding concern.

“Clinical trials are fraught with potential risks and benefits, and the goal is to minimize the risks,” says Richard Klein, director of the FDA's Patient Liaison Program in the Office of Health and Constituent Affairs (OHCA). “When there is a hold, it means the FDA is doing its job by keeping an eye on the outcomes and adverse events and stopping a trial when there's indication of a serious problem.”

When significant safety questions arise during a trial, the FDA has the option to place either a partial or full clinical hold on the trial, explains Deborah Miller, PhD, MPH, cancer patient liaison in the OHCA. In both cases, new patient enrollment is halted, but a partial hold allows enrolled patients to continue receiving the experimental treatment.

When a full hold is placed, patients already enrolled may continue therapy if the FDA approves a request for expanded access, says Miller. Such requests, which can be made by the trial sponsor or individual patients, might be granted when the benefits of continuing the drug outweigh the risks for a particular subgroup.

In cancer drug trials, patient deaths do not necessarily trigger a hold unless there is reason to believe that they were due to the investigational treatment, she notes. Regulators assume that some patients won't respond to the treatment or will die of the disease itself during the course of a trial.

However, the decision to impose a hold on a cancer trial is never easy because the stakes are so high, Miller says. Patients with fatal illnesses who have run out of treatment options are often willing to accept substantial risk. At the same time, drug companies invest extensively in getting a trial up and running and may be unable to afford lengthy delays.

“A trial has an enormous setup cost, and when a hold occurs, companies are still paying all the bills but not making any progress on the trial,” says Klein. “At the same time, they might lose patients who decide they don't want to wait and move to another therapy, which disqualifies them from the study.”

For example, Los Angeles, CA–based CytRx last summer blamed a partial clinical hold for its disappointing results from a phase III study of aldoxorubicin—a tumor-targeted doxorubicin conjugate—for the treatment of advanced soft-tissue sarcomas. The company's shares plunged by 60%.

Because the trial was being conducted under a special protocol assessment granted by the FDA, the company was required to analyze data after the disease progressed in 191 patients. The timing of the analysis did not allow for sufficient follow-up of the patients who entered the study after the hold was lifted, company officials maintain. As a result, nearly half of all patients were excluded from progression-free-survival (PFS) data in the interim analysis.

“Because enrollment was interrupted by a clinical hold, both PFS and response data need to be analyzed at a future date to account for patients enrolled later in the trial,” says Sant Chawla, MD, the trial's principal investigator and director of the Sarcoma Oncology Center in Santa Monica, CA. The company expects to complete a second analysis and meet with the FDA by year's end.

For Juno, even the short hold delayed the developmental timeline by about 2 months, says Gilbert, because investigators enrolled only one new patient per week for 6 weeks after the trial resumed to monitor for serious adverse effects.

“We're now coming up on our sixth patient, and if we do not see additional toxicity with single-agent chemotherapy, we will open up enrollment to more new patients,” he says. “We needed to understand the safety of the product in this small patient group before we could move forward.”–Janet Colwell

For more news on cancer research, visit Cancer Discovery online at http://cancerdiscovery.aacrjournals.org/content/early/by/section.