Optimizing Outcomes in Clinical Trials of CAR T-Cell Therapy for Patients With ALL

Optimizing Outcomes in Clinical Trials of CAR T-Cell Therapy for Patients With ALL

Although 60% to 80% of adults with acute lymphoblastic leukemia (ALL) achieve remission following chemotherapy, about half of them relapse after the initial treatment. For adults with relapsed or refractory ALL, the clinical outlook is grim. Five-year overall survival rates range from 14% in patients older than age 55 to 69% in patients younger than age 30.

In recent years, progress has been made in developing chimeric antigen receptor (CAR) T-cell therapy to treat relapsed and refractory ALL. The downside of CAR T-cell therapy is that it can cause a potentially fatal condition called cytokine release syndrome and neurotoxicity. Research to identify the most appropriate patient population for this therapy, and to reduce its toxicity, was highlighted in the Hematologic Malignancies—Leukemia, Myelodysplastic Syndromes, and Allotransplant Oral Abstract Session held on Saturday, June 4.

“Unprecedented Remission Rates” With CAR T-Cell Therapy

The use of CAR T-cell therapy involves the following steps: Patients undergo leukapheresis and their T cells are expanded and activated ex vivo and transduced with a gene that expresses the CAR. The T cells are then expanded and, after receiving lymphodepletion, the patients receive an infusion of the modified T cells. Much of the research, including the studies presented in the session, have looked at the effect of T cells that express a CAR that binds to CD19, which is on the surface of normal and malignant B cells; modified CAR T cells kill CD19 cells when they bind.

Dr. Noelle V. Frey
Noelle V. Frey, MD, of the Abramson Cancer Center of the University of Pennsylvania, discussed what she called “unprecedented remission rates” of approximately 70% to 90% that have been observed for various investigational CAR T-cell therapies including the CD19-directed therapy CTL019. “Unfortunately, the immune activation that is critical for these high response rates is also responsible for this therapy’s most significant treatment-related toxicity,” Dr. Frey said.

The studies that reported high remission rates for CTL019 also observed cytokine release syndrome (CRS), one of the most critical adverse events of CAR T-cell therapy. It affected between 76% and 100% of patients, and in 27% to 43% of them complications were severe. Disease burden of ALL at baseline correlates strongly with severity of CRS.

CRS is marked by increases in cytokines including interleukin (IL)-6, IL-10, C-reactive protein, and interferon gamma. The syndrome can cause mild initial symptoms such as fever, fatigue, and muscular pain and can progress to hypertension and hypoxic respiratory failure that require supportive treatment such as intravenous fluids or treatment with the IL-6 inhibitor tocilizumab.

Reducing Toxicity Associated With Therapy

Dr. Frey and her colleagues have embarked on two clinical trials, NCT02030847 and NCT01029366, involving a total of 30 adults with CD19-positive relapsed or refractory ALL. These trials will help determine whether varying the dosing and scheduling of CTL019 could reduce CRS. They found that a high-dose of CTL019, which contained 5 × 108 T cells, had a response rate of 100% but also a CRS rate of 100%. Splitting the same dose over 3 days was associated with 86% response rate and 66% CRS rate, whereas administering a single low-dose infusion, containing 5 × 107 T cells, reduced both efficacy and CRS, to 33% and 66%, respectively.

Three of six of the patients who received a single high-dose of CTL019 and developed CRS died in the days to weeks following their infusion. All of these patients were found to have infections, either with influenza, pseudomonas, or stenotrophomonas, or sepsis; these diseases, along with CRS, probably contributed to their deaths. These findings suggest that clinicians should conduct aggressive infectious disease monitoring and treat patients with antimicrobials prior to CAR T-cell therapy, Dr. Frey said.

Unlike with CRS, the trials did not find a relationship between CTL019 dose and rates of adverse neurologic events such as seizure and encephalopathy.

Dr. Frey concluded her presentation (Abstract 7002) by stating that in a split or fractionated dosing scheme, clinicians can monitor patients after each dose and delay or cancel subsequent doses if toxicity occurs, as she and her colleagues did in their recent trials. In addition, she suggested that alternatives to fractionated dosing are being investigated in clinical trials, such as varying the timing of anticytokine therapy.

Impact of Disease Burden on Treatment Outcomes

There could be further opportunities to improve CAR T-cell therapy by modifying the dosing scheme according to amount of disease at baseline, and possibly testing the therapy earlier in disease stage. Jae H. Park, MD, of Memorial Sloan Kettering Cancer Center, presented data from a phase I clinical trial he is leading, NCT01044069, to explore the relationship between disease burden prior to infusion with remission rates in 51 adults with relapsed or refractory ALL (Abstract 7003).

The trial is using a CAR called CD19-28z. Similar to the CAR in the two trials that Dr. Frey presented, CD19-28z directs T cells to CD19-expressing cells; however, it contains an intracellular domain from CD28, which could improve the expansion and persistence of CAR T cells in patients.

To determine the baseline disease burden, Dr. Park and his colleagues measured blasts in the bone marrow before infusion and categorized morphologic disease as having at least 5% blasts whereas minimal disease was fewer than 5% blasts. Based on earlier research finding that patients with more disease burden had more side effects following treatment, they decided to give patients with morphologic disease a lower dose of 1 × 106 CAR T cells per kilogram, whereas patients with minimal disease received 3 × 106 CAR T cells per kilogram.

For the last nine patients they treated in the study, Dr. Park and his colleagues decided to administer conditioning chemotherapy 2 days prior to the infusion of T cells.

Treatment with CD19-28z achieved high rates of complete remission (CR) and CR that was negative for minimal residual disease (MRD) regardless of the amount of disease at baseline. Among the patients with morphologic disease, 77% achieved CR by 20 days after treatment, and 90% of patients with CR were negative for MRD. The CR rate among patients with minimal disease at baseline was 90% by 25 days after treatment, and 78% of the patients were negative for MRD. The presence of MRD in patients with CR appeared clinically important, as 45% of these patients relapsed.

However, the disease burden at baseline did have an important impact on toxicity associated with treatment, as patients with minimal disease went on to experience significantly less CRS and neurotoxicity compared with patients with morphologic disease.

Remaining Questions for CAR T-Cell Therapy

In his remarks following the presentations, Ehab L. Atallah, MD, of the Medical College of Wisconsin, called the success of CAR T-cell therapy amazing. Even though the concept has been around since the 1980s, the study of this possible treatment option has exploded recently, with groups in the United States, Europe, China, Japan, and Australia all conducting this therapy.

However, numerous questions remain about the clinical use of CAR T-cell therapy, Dr. Atallah said. For example, what are the best CAR constructs and how can we optimize steps in this treatment, such as lymphodepletion? In addition, Dr. Atallah said, the data from the trial by Dr. Park and his colleagues suggest that it is perhaps time to move that therapy upfront rather than waiting for relapse.

Carina Storrs