High-dose lymphodepletion regimen effective but more toxic for patients receiving TILs
Click Here to Manage Email Alerts
A lower dose of cyclophosphamide during preconditioning prior to tumor-infiltrating lymphocyte infusion appeared comparably effective but less toxic than a higher dose, according to results published in Journal for ImmunoTherapy of Cancer.
Results showed two regimens — 120 mg/kg cyclophosphamide and 125 mg/m2 fludarabine, or 60 mg/kg cyclophosphamide and 125 mg/m2 fludarabine — conferred similar benefit for bone marrow depletion and recovery.
Patients assigned the lower cyclophosphamide dose experienced fewer cardiotoxic effects and shorter duration of hospitalization. However, no patients who received the lower dose of cyclophosphamide achieved objective response to tumor-infiltrating lymphocyte (TIL) therapy.
Certain inflammatory biomarkers after the start of nonmyeloablative lymphodepletion may help predict response and OS after TIL therapy, results showed.
“A common feature of all current T-cell therapies — no matter if they use gene-engineered T cells or TIL — is the non-myeloablative (NMA) lymphodepleting preconditioning of patients prior to cell infusion,” Abraham Nissani, MD, professor of surgery and head of the department of general and oncology surgery at Sheba Medical Center in Israel, and colleagues wrote. “Although there is a common consensus that NMA preconditioning is required to obtain a durable objective response in patients with cancer treated with adoptive T-cell therapy, there is no study comparing the different NMA lymphodepleting regimens head-to-head.”
Nissani and colleges analyzed data from three phase 2 trials of adoptive cell therapy conducted at Sheba Medical Center.
Patients with stage IV melanoma received TILs as salvage therapy, and patients with relapsed or refractory acute lymphoblastic leukemia received CD19-directed chimeric antigen receptor T-cell therapy.
Patients with melanoma underwent nonmyeloablative lymphodepletion with one of three cumulative-dose regimens: 120 mg/kg cyclophosphamide and 125 mg/m2 fludarabine; 60 mg/kg cyclophosphamide and 125 mg/m2 fludarabine; or total body irradiation and 75 mg/m2 fludarabine.
Patients with ALL underwent a preconditioning regimen of 30 mg/kg cyclophosphamide and 75 mg/m2 fludarabine.
Patients in the three melanoma cohorts had similar characteristics except for prior treatments received. Those who received the higher-dose cyclophosphamide/fludarabine regimen received treatment before the use of immune checkpoint inhibitors, whereas those who received the lower-dose regimen had experienced disease progression after immune checkpoint inhibitor therapy.
Nissani and colleagues aimed to evaluate the effect of each lymphodepletion regimen on safety and treatment efficacy. Researchers collected data on bone marrow suppression and recovery, serum cytokine levels, clnical responses and treatment-related adverse events.
Analysis of severe lymphodepletion on the day of TIL infusion as measured by absolute lymphocyte counts showed similar values when comparing the high-dose and low-dose cyclophosphamide/fludarabine groups (0.029 ± 0.036 K/L [median, 0.020 K/L] vs 0.024 ± 0.017 K/L [median, 0.025 K/L]).
All eight patients in the low-dose cyclophosphamide/fludarabine group achieved severe neutropenia, defined as absolute neutrophil count of 0.5 K/L or less. Similarly, nearly all of the 103 patients in the high-dose group achieved severe neutropenia (98%; n = 101).
The results showed both regimens “were equally efficient in achieving deep lymphopenia and neutropenia,” according to the researchers.
Patients in the total body irradiation group did not achieve severe neutropenia and achieved only “a minor decrease” in absolute lymphocyte counts, researchers wrote.
Bone marrow recovery appeared similar between the high- and low-dose cyclophosphamide/fludarabine groups, with patients in each cohort recovering from severe neutropenia by day 6 or day 7 after TIL infusion. Patients in all cohorts had recovery of absolute lymphocyte counts within normal ranges within 4 to 7 days after infusion.
Twenty-eight percent of patients in the high-dose cyclophosphamide/fludarabine cohort responded to TIL therapy. No patients in the low-dose cyclophosphamide/fludarabine cohort or the total body irradiation cohort responded to TIL therapy.
Researchers hypothesized that the lack of objective responses to TIL therapy in the low-dose cyclophosphamide/fludarabine cohort likely was due to all patients being refractory to previous treatment with anti-PD1 therapy.
“Interestingly, among the 17 anti-PD-1-refractory patients in the [higher-dose] cohort, 16 patients did not respond to TIL therapy,” Nissani and colleagues wrote. “[This finding] strengthens our hypothesis that patients refractory to anti-PD-1 antibodies are less likely to respond to TIL therapy compared with anti-PD1 naive patients.”
Researchers evaluated patients in the high-dose cyclophosphamide/fludarabine group to assess predictive biomarkers for clinical response.
Results showed patients who responded to TIL therapy had significantly higher absolute lymphocyte counts by day 4 after the start of nonmyeloablative lymphodepletion (P .001).
A significantly higher percentage of patients who achieved objective response to TIL therapy had post-lymphodepletion neutrophil-to-lymphocyte ratios of 1.79 or less (61% vs. 25%; P = .002). Similarly, a higher percentage of patients who responded to therapy had a post-lymphodepletion platelet-to-neutrophil ratio of 32.7 or less (61% vs. 15%; P .001).
Kaplan-Meier analysis showed patients in the high-dose group with lower values of both ratios achieved significantly longer survival.
Median OS was 23 months for patients with a neutrophil-to-lymphocyte ratio of 1.79 or less vs. 13.5 months for those with a higher ratio (P = .011). Median OS was 39 months for patients with a platelet-to-neutrophil ratio of 32.7 or less vs. 12.5 months for those with a higher ratio (P .001).
Three deaths from cyclophosphamide-induced cardiotoxicity occurred in the high-dose cyclophosphamide/fludarabine group; none occurred in the low-dose group.
Eighty-eight percent of patients in the high-dose cyclophosphamide/fludarabine group required red blood cell/platelet transfusions to prevent anemia or related complications compared with three of eight (37.5%) patients in the low-dose group.
Patients in the low-dose cyclophosphamide/fludarabine group spent significantly less time hospitalized after nonmyeloablative lymphodepletion than those in the high-dose group (15.4 ± 1.5 days vs 20.8 ± 5.1 days; P = .004).
“A major aspect that must be taken into account when choosing chemotherapy for lymphodepletion is the toxicity related to it,” Nissani and colleagues wrote. “Patients in [the high-dose] cohort developed significantly more acute adverse events, including thrombocytopenia, febrile neutropenia and cardiotoxicity.”
Perspective
Back to Top
Yago L. Nieto, MD, PhD
Lymphodepleting chemotherapy preceding the infusion of adoptive cellular immunotherapy — such as CAR T cells, CAR-natural killer cells or TILs — is necessary to optimize the expansion, persistence and activity of the infused cells.
Nissani and colleagues reported on their experience with lymphodepletion followed by TILs among patients with metastatic melanoma. Their observation that lymphocyte recovery with higher absolute lymphocyte counts correlates with response is of interest and adds to the wealth of data on the prognostic effect of this marker of early immunological recovery in other settings, including after CAR T-cell therapy for ALL or allogeneic or autologous hematopoietic stem cell transplant for various malignancies.
Additionally, the researchers performed a cross-trial comparison of three lymphodepleting regimens. Regimens comprising 120 mg/kg cyclophosphamide and 125 mg/m2 fludarabine and 60 mg/kg cyclophosphamide and 125 mg/m2 fludarabine resulted in more favorable hematologic responses — initial neutropenia and lymphopenia, both of which should favor the expansion of the infused immune cells, and absolute lymphocyte counts at subsequent recovery — than the total body irradiation regimen.
However, the 120 mg/kg cyclophosphamide and 125 mg/m2 fludarabine regimen resulted in 3% of early deaths from cyclophosphamide-related cardiomyopathy, which prompted the investigators to abandon the regimen.
Researchers only observed tumor responses after TILs only among those who underwent lymphodepletion with the higher-dose cyclophosphamide combination, although the patient populations might have been unbalanced across the three cohorts.
Perhaps the lower-dose cyclophosphamide regimen used by the investigators could be optimized. For example, given the long serum half-life of fludarabine (20 hours), daily administration of this drug until day –1 likely had a deleterious effect on the TIL product infused the next day. Thus, to the authors’ conclusion that toxicity should influence the selection of the lymphodepleting regimen, one might add that careful consideration of chemotherapy scheduling is just as important.
Collapse
Click Here to Manage Email Alerts