乌鸦传媒

ATLANTA -- After a long string of clinical trial success in hematologic malignancies, chimeric antigen receptor (CAR) T-cell therapy failed to improve event-free survival versus standard of care for primary refractory/relapsed aggressive B-cell non-Hodgkin lymphoma.

After a median follow-up of 10 months in the international phase III , patients treated with tisagenlecleucel (tisa-cel, Kymriah) and those randomized to standard care had identical median event-free survival (EFS) of 3.0 months. Adjustment for potential imbalances in baseline characteristics did not substantively affect the results, which showed a hazard ratio of 0.95 for tisa-cel versus standard therapy (95% CI 0.72-1.25).

The explanation for the unexpected outcome "is a question that keeps me up every night," said Michael R. Bishop, MD, of the University of Chicago, during a late-breaking abstraction presentation at the American Society of Hematology (ASH) annual meeting.

He cited several possible answers: differences in trial design (including two positive studies reported at ASH), a liberal policy for use of bridging chemotherapy in BELINDA, substantial differences in time to infusion of CAR-T therapy, and different strategies for lymphodepleting therapy.

Responding to a question from the audience, Bishop said, "As you mentioned, tisa-cel has been very successful in other lines of diffuse large B-cell lymphoma, but I think our trial design did not permit it to have ultimate efficacy."

The results were reported simultaneously in the (NEJM).

Transformative Therapy

CAR T-cell therapy burst into the therapeutic landscape with two closely timed FDA approvals in the second half of 2017. The approvals stipulated later-line use of tisagenlecleucel and axicabtagene ciloleucel (axi-cel, Yescarta) in progressive leukemia and non-Hodgkin lymphoma (NHL), respectively. Since then, clinical investigators have pushed the envelope for earlier use to treat aggressive hematologic malignancies that many specialists believe will ultimate prove refractory and become fatal in most cases.

In two randomized trials reported at ASH, axi-cel and lisocabtagene maraleucel (liso-cel, Breyanzi) significantly improved EFS and progression-free survival (PFS), respectively, versus standard second-line therapy for progressive NHL (chemotherapy with or without stem-cell transplantation). Data from showed that incorporating CAR T-cell therapy into first-line treatment of aggressive large B-cell lymphoma (LBCL) led to objective responses in 89% of patients, including complete responses (CRs) in 78%.

The BELINDA trial involved 322 transplant-eligible adults with primary refractory NHL or disease that had relapsed within 12 months of first-line treatment. The patients were randomized to tisa-cel or standard care chemotherapy. Patients allocated to CAR T-cell therapy could receive bridging therapy as needed to maintain disease control until T-cell infusion.

Patients who achieved a partial or complete response with standard-of-care therapy received additional chemotherapy, followed by high-dose chemotherapy and autologous stem-cell transplant. Patients who had stable or progressive disease with chemotherapy received additional chemoimmunotherapy, and selected patients were offered CAR T-cell therapy within constraints of the production process.

The primary endpoint was EFS, as assessed at 12 weeks. Secondary endpoints included objective response, safety, and cellular kinetics.

Before presenting the outcome data, Bishop noted the variable but generally prolonged time to infusion of CAR T-cell therapy. Median time to infusion was 52 days overall, including 41 days for U.S. patients and 57 days for patients outside the U.S. Two-thirds of patients enrolled in BELINDA were from outside the U.S. In contrast, the successful U.S.-based ZUMA-7 randomized trial reported at ASH had a median time to infusion of 27 days.

Almost half (48%) of the patients randomized to tisa-cel received more than one cycle of bridging chemotherapy, and only 17% received no bridging therapy, an indication of the prolonged time to CAR T-cell therapy infusion. Additionally, the overall response at 6 weeks was 38% in the tisa-cel arm and 54% in the control group. Almost twice as many patients in the tisa-cel arm had progressive disease at 6 weeks, prior to T-cell infusion (25.9% vs 13.8%). By 12 weeks, the response rates were 46% with tisa-cel and 42.5% with standard therapy.

The two treatment arms had similar rates of adverse events (AEs), treatment-related AEs, and serious AEs. Cytokine release syndrome (CRS) occurred in 58.6% of patients treated with tisa-cel, but only 4.9% of patients had grade ≥3 CRS. Neurologic events occurred in 10.3% of the tisa-cel arm, including grade ≥3 in 1.9%.

On-study mortality was 32.1% in the tisa-cel arm and 28.1% in the control group. Most deaths in both arms resulted from progressive disease, said Bishop. AE-related mortality was 6.2% with tisa-cel and 8.1% with the control treatment.

"Effective bridging therapy prior to CAR T-cel infusion and a shorter time to infusion for this chemotherapy-refractory patient population could be critical to improve outcomes," said Bishop. "Insights from this randomized phase III study should help guide optimal use of CAR T-cells in patients with relapsed/refractory aggressive non-Hodgkin lymphoma requiring second-line therapy and design of future CAR-T trials."

What Went Wrong?

Any suggested explanation for the study's negative outcome "would be pure speculation," said Frederick L. Locke, MD, of Moffitt Cancer Center in Tampa, Florida, who reported the successful ZUMA-7 results.

"One thing that's remarkable to me is that if you look at the standard-of-care arms in all three of these studies, they seem to be remarkably similar in terms of event-free survival," he told 乌鸦传媒. "There are different rates of patients who got to transplant, but the ongoing event-free survival in all three studies with the standard of care was poor. For whatever reason, it appears that the CAR T-cell arm didn't perform as well as expected."

The disparate results from the trials reported at ASH show that patient selection for CAR T-cell therapy is important, said the authors of an accompanying NEJM .

"What has been learned is that not all patients with 'relapsed or refractory large B-cell lymphoma' are the same," wrote Wyndham H. Wilson, MD, PhD, and Mark Roschewski, MD, of the Lymphoid Malignancies Branch of the National Cancer Institute, and NEJM deputy editor Dan L. Longo, MD. "Assessment of the curative potential of different constructs of CAR T cells will require comparisons in similar groups of patients, and differences in curative potential cannot be ruled out as contributing factors for the different outcomes of these trials."