Efficacy and safety of liso-cel in patients with NHL and CLL/SLL – update from ASH 2019 annual meeting

Chimeric antigen receptor (CAR) T-cell therapy is evolving the treatment landscape of B-cell non-Hodgkin lymphoma (NHL). Currently, two CAR T-cell therapies, axicabtagene ciloleucel and tisagenlecleucel, are approved for the treatment of relapsed/refractory (R/R) lymphoma. A third anti-CD19 CAR T-cell product, in advanced stages of clinical development, is lisocabtagene maraleucel (liso-cel; also known as JCAR017), which is administered as a defined composition of CD4⁺/CD8⁺ CAR T cells.^1,2

Liso-cel was granted breakthrough therapy designation in R/R, aggressive large B-cell NHL, by the U.S. Food and Drug Administration (FDA) and an orphan designation by the European Commission for the treatment of follicular lymphoma (FL) and diffuse large B-cell lymphoma (DLBCL).

During the 61^st American Society of Hematology (ASH) Annual Meeting & Exposition, several oral and poster sessions presented data from the ongoing clinical trials evaluating the safety and efficacy of liso-cel in patients with NHL and chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL/SLL). Here we provide a summary of the data reported at the meeting. For the oral sessions that the article is based on, data presented at the ASH meeting may supersede the data in the published abstract.

TRANSCEND NHL 001

TRANSCEND NHL 001 (NCT02631044) was an open-label, multi-center phase I clinical trial evaluating the safety and pharmacokinetics of liso-cel in adult patients with R/R B-cell NHL. 

Abstract #241: Pivotal safety and efficacy results³

Key safety and efficacy results in patients with R/R large B-cell lymphomas (LBCL), were presented by Jeremy Abramson, Massachusetts General Hospital Cancer Center, US.

Trial eligibility: adult patients with DLBCL not otherwise specified (NOS; including transformed from any indolent lymphoma), high-grade B-cell lymphoma (HGBCL) with MYC and BCL-2 and/or BCL-6 rearrangements, primary mediastinal B-cell lymphoma (PMBCL), or FL grade 3B (FL3B). Patients had R/R disease after a minimum of two prior lines of therapy. Patients with grade 3/4 cytopenias, mild-moderate organ dysfunction, and secondary CNS lymphoma were eligible. Bridging therapy was allowed after leukapheresis. Patients with PET-positive disease underwent lymphodepletion with fludarabine (30 mg/m²) and cyclophosphamide (300 mg/m²) for 3 days. Liso-cel was infused 2–7 days after lymphodepletion.

The study was conducted in three phases: dose finding (n= 60 patients), dose expansion (n= 83) and dose confirmation (n= 126). In the dose-finding phase of the study, patients were enrolled at one of three target dose levels (DL): DL1: 0.5×10⁸, 1×10⁸, or 1.5×10⁸ viable CAR T cells. All dose levels were then expanded and DL2 was chosen for confirmation cohort. Primary endpoints were treatment-emergent adverse events (TEAEs) and overall response rate (ORR) by an independent review committee (IRC). Secondary endpoints included complete response (CR) rate, duration of response (DOR), progression-free survival (PFS), PFS ratio, and overall survival (OS).

Results

In total, 344 patients were leukapheresed. Out of these patients, 50 did not receive liso-cel (predominantly due to disease progression or death). Out of 294 patients who received liso-cel, 25 patients received nonconforming product. There were 269 patients with LBCL who received liso-cel conforming to standards (DL1, n= 51; DL2, n= 177; DL3, n= 41) and 256 patients were included in the efficacy set (DL1, n= 46; DL2, n= 169; DL3, n= 41). Patient baseline characteristics are shown in Table 1 and it is worth noting that 89% of patients had high-risk features.

The overall safety profile was deemed acceptable (Table 2). Nineteen patients (7%) required admission to intensive care unit and seven patients (3%) experienced grade 5 TEAEs, four out of which were considered to be treatment-related (diffuse alveolar damage, pulmonary hemorrhage, multiple organ dysfunction syndrome, and cardiomyopathy) and three unrelated to liso-cel (fludarabine leukoencephalopathy, septic shock, and progressive multifocal leukoencephalopathy). No grade 5 cytokine release syndrome (CRS) or neurological events (NE) were recorded. In total, 42% of patients experienced CRS, including 2% of patients with grade 3/4. The median time to both onset and resolution was 5 days. NEs, including grades 3 and 4, were observed in 30% of patients, (9% and 1% of patients, accordingly) and the median time to onset was 9 days, with a median of 11 days to resolution.

Other adverse events (AEs) of special interest included prolonged grade ≥ 3 cytopenias (in 37% of patients), grade ≥ 3 infections (12%), grade 1/2 infusion-related reactions (1%), ≥ 3 tumor lysis syndrome (1%), and grade 1/2 hypogammaglobulinemia (14%).

After the median follow-up of 12 months (95% CI, 11.2–16.7), liso-cel demonstrated high response rates across different subgroups of patients with LBCL (Table 3). The median PFS and OS were 6.8 months and 21.1 months accordingly but were not reached in patients with CR (after a median follow-up of 12.3 months and 17.6 months, respectively). Subgroup analysis by histology revealed higher PFS in patients with PBMCL and transformed FL compared with double or triple hit NHL, DLBCL NOS, or HGBCL. Patients with comorbidities had inferior PFS compared with those without (3 vs 9.5 months; HR= 1.5, CI 95%, 1.0–2.2; p= 0.03). Patients who had received bridging therapy experienced a shorter PFS versus those patients who had not (5 vs 14.1 months; HR= 1.3, CI 95%, 0.9–1.9; p= 0.13).

Analysis of cellular kinetics showed similar levels of expansion of CD4⁺ and CD8⁺ CAR+ T cells, with the median peak expansion 11 days after infusion. Moreover, CAR T-cells were detectable for one year in just over half of patients (n= 37/70).

In conclusion, this largest clinical study to date of anti-CD-19 CAR T product in patients with LBCL demonstrated a rapid and high-rate of durable response with manageable toxicity profile in patients with high-risk prognostic factors and across different histologic subtypes. Authors suggest that the low incidence and delayed onset of serious TEAEs with liso-cel raise the possibility of treatment in the outpatient settings. A poster with initial data on the use of liso-cell in the outpatient clinic was also presented⁴ and is summarized in the last section of this article.

UPDATE: The results of the study were published in Lancet Haematology in September 2020, and can be accessed here.

Abstract #66: Patient-reported outcomes⁵

Patient-reported outcomes (PRO) are important when evaluating the experience of patients receiving treatment and can influence treatment selection decisions and reimbursement. David G. Maloney from Fred Hutchinson Cancer Research Center, US, presented an abstract on quality of life and health utility in patients with R/R aggressive B-cell NHL enrolled in the TRANSCEND NHL 001 study. The data incorporated PRO assessments to evaluate the impact of liso-cel on symptoms, health-related quality of life (HRQoL), and health utility in patients with high disease burden and toxicities associated with prior chemotherapy and HSCT.

Primary assessments included global health status, physical functioning, fatigue, and pain by European Organisation for Research and Treatment of Cancer (EORTC) quality of life questionnaire core 30 (QLQ-C30), and health index score by EuroQol 5-dimensions 5-level questionnaire (EQ-5D-5L). Exploratory assessments of role, cognitive, emotional, and social functioning were done using QLQ-C30, and imaginable health status by EuroQol visual analogue scale (EQ-VAS). The assessments were completed at screening, baseline, 1–3, 6, 9, 12, 18, and 24 months, at disease progression, and the end of the study.

The survey participation rates were high (81/128 patients had at least 6 months follow-up and 38/51 had at least 12 months follow-up). Median on-study follow-up was 8.7 and 8.8 months for patients in the QLC-C30 (n= 181) and EQ-5D-5L (n= 186), respectively.

Analysis of EORTC QLQ-C30 showed clinically meaningful (≥ 10 points difference) improvement in global health status (16.9 points median change from baseline) and reduced fatigue (-15.5 points difference from baseline) with liso-cel treatment, as early as 1 month and through 12 months. However, no clinically meaningful differences were observed in physical functioning (6.5 points median change from baseline) or pain (-8.8 points median change from baseline) over the same time. Additionally, clinically meaningful improvements in HRQoL and symptom burden were greater in patients experiencing clinical response to liso-cel. Liso-cel also improved health state index score (EQ-5D-5L assessment) and visual analog scale (EQ-VAS) from the 3^rd month through to month 12.

Altogether, the results demonstrate significant improvement in patients’ HRQoL and symptom burden with liso-cel therapy.

Abstract #202: Patterns of immune infiltration associated with durable response⁶

Various tumor and patient related factors can influence response to CAR T treatment. Authors of the next TRANSCEND NHL 001 oral abstract aimed to gain a better understanding of how factors in the tumor microenvironment (TME) can impact the responses to liso-cel. The data were presented by David Reiss from Bristol-Myers Squibb.

In total, 111 biopsies from 83 patients in the DLBCL patient cohort (58 collected at baseline and 53 around 11 days after liso-cel infusion), including 28 paired, were analyzed by multiplexed immunofluorescence imaging and RNA sequencing.

Analysis of pre-treatment biopsies revealed that patients who achieved an early CR had a higher level of PD-1⁺ T cells (p= 0.0085) and showed a trend (p= 0.12) to have elevated levels of total T cells (CD4⁺ and CD8⁺) activation compared with patients who experienced PD. Patients with durable CR were characterized by presence of fewer immunosuppressive macrophages, in particular PD-L1⁺ (p= 0.02).

In post-treatment biopsies, infiltration of CAR T cells was greater in patients with early and durable CR (approximately 5% of total T cells), as well as increased number of total (mainly CD4⁺, p= 0.0081)) and GZMB⁺ endogenous T cells (p= 0.031 and p= 0.0069) between pre-treatment and Day 11 biopsies. A similar increase was also seen in immunosuppressive macrophages (p= 0.026), including PD-L1⁺ (p= 0.001) and IDO1⁺ (p= 0.00025) in patients with CR.

The results confirmed that the TME impacts on responses to liso-cel in patients with DLBCL. Patients with a lower number of immunosuppressive macrophages and higher infiltration of active T cells, in the TME show superior infiltration of CAR T cells after infusion with liso-cel and are more likely to achieve durable CR. Both endogenous T cells (CD4⁺ in particular) and CAR T-cells play an important role in generating anti-tumor immune response leading to CR in those patients.

TRANSCEND CLL 004

Although the use of novel combinations and the introduction of targeted therapies, such as ibrutinib and venetoclax, are improving outcomes for patients with CLL, they are not curative and have inadequate rates of CR and undetectable measurable residual disease (uMRD).^7,8 Over time, patients relapse with refractory disease and have poor outcomes.^9,10 Therefore, effective treatment options for those patients are desperately needed.

TRANSCEND CLL 004 (NCT03331198), a phase I/II clinical trial, is evaluating the safety and efficacy of liso-cel in heavily pre-treated patients with R/R CLL or SLL, including patients previously treated with ibrutinib, a BTK inhibitor (BTKi), and venetoclax, a BH3 mimetic.

Abstract #503: Updated safety and efficacy results¹¹

Tanya Siddiqi from City of Hope, US, reported updated results from the study, summarized below. The Lymphoma Hub also had a chance to speak with her and you can watch the interview, where she describes main findings from the study of this CAR T-cell therapy.

Patients with R/R CLL/SLL, who failed or were ineligible for BTKi therapy, with at least two prior therapies, were eligible for the study. After enrolment, patients underwent leukapheresis (bridging therapy was allowed during the liso-cel manufacturing) and had measurable disease confirmed, before lymphodepletion with fludarabine and cyclophosphamide for 3 days, followed by liso-cel infusion 2–7 days later. Liso-cel manufacturing success rate was 96%.

There were 23 evaluable patients in the dose-escalation study (nine patients at DL1 of 0.5×10⁸ viable CAR⁺ T cells and 14 patients at DL2 of 1×10⁸ viable CAR⁺ T cells), which was conducted over 28 days. The primary endpoints were safety and establishing the recommended dose, while anti-tumor activity and pharmacokinetic profile were exploratory endpoints. Baseline patient characteristics are described in table 4.

The safety profile (Table 5) was assessed in all 23 patients. Apart from a slight increase in CRS incidence, the safety profile of liso-cel in patients with CLL/SLL was similar to that described above in patients with LBCL (Table 2). Two dose-limiting toxicities were recorded at DL2 but resolved with appropriate management, one patient had grade 4 hypertension and the other patient had grade 3 encephalopathy, grade 3 muscle weakness, and grade 4 tumor lysis syndrome. Any grade CRS occurred in 17 patients (74%) with the median time to first onset of 4 days (1–10) and a median duration of 12 days (2–50). NEs of any grade were reported in nine patients (39%), with the median time of 4 days (2–21) to the first onset, and the median duration of 21 days (6–56). There were no reports of grade 5 CRS or NE.

Patients experienced high levels of response and uMRD (table 6) with liso-cel, with 68% of patients responding by Day 30 and response deepened over time in 27% of patients. Moreover, 83% (10/12) of patients who achieved PR or CR at 6 months remained in response at 9 months, including 8 patients in response at 12 months or longer. Similar, maintenance of response was seen with uMRD, with 80% (12/15) of patients continuing to respond.

Pharmacokinetics and pharmacodynamics assessment revealed a rapid elimination of CD19⁺ B cells, with 87% of patients demonstrating long-term suppression and 80% of patients maintaining CAR⁺ T cells at 6 months. The investigators also found higher lymph node tumor burden, as well as elevated levels of IL-16, and TNFα to be associated with NEs.

In summary, liso-cel demonstrated a manageable toxicity profile and encouraging efficacy in heavily pre-treated patients with high-risk CLL/SLL, including patients who were resistant or refractory to ibrutinib and venetoclax. Responses were rapid, durable and improved over time. The phase II portion of the trial is currently enrolling at DL2 and will further evaluate the toxicity and efficacy in this subset of patients. 

Other liso-cel data presented at ASH 2019

In addition to the oral abstracts described above two posters were also presented during the ASH meeting, one describing initial results from the PILOT study and second one with the data combining on outpatient treatment of liso-cel in patients with R/R B-cell NHL, including transplant non-eligible from TRANSCEND NHL 001, OUTREACH, and PILOT studies.

Poster # 2882: TRANSCEND-PILOT¹²

In contrast to TRANSCEND NHL and TRANSCEND CLL/SLL studies, which investigate liso-cel as the third line of treatment, TRANSCEND PILOT (NCT03483103) is evaluating this CAR T product as second-line therapy for patients with R/R aggressive B-cell NHL ineligible for HSCT. Alison R. Sehgal, University of Pittsburgh Medical Center, US, and colleagues reported preliminary data from the ongoing phase II trial, which is the first to evaluate CAR T therapy exclusively in this patient population.

Adult patients had lymphodepletion, followed by infusion of liso-cel (1×10⁸ CAR⁺ T cells). At the data cut-off, therapy was successfully manufactured and administered to nine patients, including five in the outpatient settings. The median age was 71 years (64–79) and most patients were male (55%). Seven patients had DLBCL NOS and two transformed FL. The cohort included two patients with triple-hit lymphoma, one transformed from FL, and four patients with a high tumor burden. Five patients relapsed on prior therapy and four had refractory disease. The median hematopoietic cell transplantation comorbidity index score was 3 (0–3).

Six patients had ≥ 1 TEAEs grade ≥ 3, mainly cytopenias, including three with prolonged cytopenia at Day 29. Infections, below grade 3, were reported in two patients. There were no reports of CRS, NE, tumor lysis syndrome, infusion reactions, grade 5 TEAEs, or treatment with tocilizumab, corticosteroid, or vasopressors. None of the outpatients required hospital admission within 29 days of liso-cel infusion. The median follow-up was 3.5 months and all treated patients responded, including four with CR (status maintained), five with PR (two maintained). The responses of patients managed in the inpatient and outpatient settings were similar. The median time to peak expansion of CAR T⁺ cells was 10 days (7–21).

The initial data demonstrated promising safety and efficacy of liso-cel as a second-line treatment for patients with R/R aggressive B-cell NHL ineligible for high-dose chemotherapy and transplantation. However, further data with longer follow-up are required.

Poster #2868: Liso-cel as an outpatient treatment⁴

The low incidence of CRS and neurologic events associated with liso-cel therapy, as well as the delayed onset, described above, potentially may allow the administration and monitoring of patients in an outpatient setting. Moving away from the inpatient treatment at university medical centers would make CAR T-cell therapy more accessible. To test the safety of this approach, Carlos R. Bachier, Sarah Cannon Blood Cancer Network, US, and colleagues combined data on the use of liso-cel in outpatient settings from the three clinical trials, TRANSCEND NHL 001, PILOT, and OUTREACH (NCT03744676). The OUTREACH is an open-label, multi-center phase II clinical trial evaluating the safety and efficacy of liso-cel therapy for patients with R/R B-cell NHL administered and monitored in the outpatient setting.

After lymphodepletion, liso-cel was administered at 0.5×10⁸, 1×10⁸, or 1.5×10⁸ CAR⁺ T cells for patients recruited on the TRANSCEND NHL 001, or 1×10⁸ CAR⁺ T cells on PILOT and OUTREACH studies. Outpatient treatment was at the discretion of the investigator. However, patients were required to have a caregiver for 30 days after liso-cel infusion, receive safety-monitoring training, and stay within one- hour travel to the site of care. CRS and treatment-related NEs were managed at a hospital.

At data cut-off, 37 patients received liso-cel and were monitored as outpatients (Table 7). Fifteen (41%) patients did not require hospitalization in the first 29 days after infusion. There were 22 patients (59%) who required hospitalization (three admitted for CRS on Day 3 or earlier, with one patient admitted to an intensive care unit for 3 days). The median time to hospitalization after treatment was 5 days (2‒22) and the median length of stay was 6 days (2‒23). Cytopenias accounted for the majority of grade ≥ 3 TEAEs (neutropenia 43%, anemia 30%, thrombocytopenia 14%). CRS was reported in 16 patients and NE in 12 patients, including two severe reversible events (Table 7). CRS and/or NEs required administration of corticosteroids alone to four patients and in combination with tocilizumab in three patients. The majority of patients across all the studies responded (Table 7). The median time to peak CAR⁺ T cell expansion was 10 days (3–21 in TRANSCEND, 7–10 in OUTREACH, and 7–10 in PILOT).

The presented data indicate the feasibility of liso-cel administration in the outpatient settings as not all patients required hospitalization. Data with longer follow-up are needed to confirm the finding.