Outcomes of Tisagenlecleucel in Lymphoma Patients With Predominant Management in an Ambulatory Setting

Abstract

Relapsed or refractory non-Hodgkin lymphoma has benefited from incorporation of chimeric antigen receptor T-cell therapy. While these adoptive immunotherapy options are revolutionary in management, it requires significant vigilance and resources to administer in an ambulatory setting. We investigated 72 patients administered tisagenlecleucel in our practice to describe their clinical outcomes and highlight our approach to outpatient management.

Introduction:

Chimeric antigen receptor T-cell therapy (CAR T) is a revolutionary adoptive immunotherapy approach in lymphoma; however, substantial resources are necessary for administration and care of these patients. Our institution has administered tisagenlecleucel primarily in an outpatient setting, and here we report our clinical outcomes.

Patients and Methods:

We conducted a single institution, retrospective study investigating outcomes of adult lymphoma patients treated with commercial tisagenlecleucel between 10/2017 and 12/2020. We analyzed patient characteristics and outcomes of efficacy and safety including overall response rate, progression-free survival, overall survival and cytokine-release syndrome, neurotoxicity, and hospitalizations.

Results:

Seventy-two patients with relapsed or refractory non-Hodgkin lymphoma (NHL) who received commercial tisagenlecleucel were identified; 68 (94.4%) patients received outpatient tisagenlecleucel. The overall response rate was 43% with a complete response observed in 25 patients (34.7%). At a median follow-up of 9.1 months, the median progression-free survival was 3.3 months. Grade 3-4 cytokine release syndrome was not observed in the study group and two patients had grade 3-4 neurotoxicity. Twenty-six patients (36.1%) were admitted within 30 days after infusion with a median length of stay of 5 days. Fourteen patients (19.4%) were admitted within 72 hours of infusion. No patient died of CAR T cell-related toxicity.

Conclusion:

Our experience affirms treatment with tisagenlecleucel in the outpatient setting is safe and feasible with close supervision and adequate institutional experience. After infusion, adverse events were manageable and the majority of patients did not require hospitalization.

Introduction

Diffuse large B-cell lymphoma (DLBCL) is the most common lymphoma subtype worldwide, accounting for 30% to 35% of cases. In the United States there are 5.6 per 100,000 new cases of large B cell lymphoma per year.^{1 , 2} While primary chemo-immunotherapy with R-CHOP can be effective in producing long-term complete remissions in up to 65% of patients with DLBCL, the remaining patients can have disease which is highly refractory. Only 20% to 25% of these relapsed and refractory patients can expect long term remission with standard salvage therapies including autologous stem cell transplantation (SCT). Those ineligible for high-dose chemotherapy have a predicted 2-year survival of 10%.^3-6 Despite progress in the development of novel agents,^7-14 these patients face an unmet medical need.

Chimeric Antigen Receptor T-cell (CAR T) therapy is a revolutionary adoptive immunotherapy approach for patients with refractory non-Hodgkin lymphoma. The JULIET, TRANSCEND and ZUMA-1 trials demonstrated sustained remission 30% to 40% of treated patients^15-19 leading to the US FDA approval of 3 different CAR T products for patients with DLBCL after ≥2 prior lines of therapy: axicabtagene ciloleucel (axi-cel), tisagenlecleucel (tisa-cel) and lisocabtagene maraleucel (liso-cel).^{15, 18, 20}

An episode of care for CAR T cells comes at high cost and resource utilization. The drug cost alone is $350,000 to $470,000.^{21, 22} In addition, many patients are admitted to scarce and expensive inpatient units for lymphodepletion, infusion and subsequent monitoring adding significantly to these costs; the median hospital stay has been reported to be 17 days.²³ The COVID pandemic additionally increased the pressure to limit elective hospitalizations for oncology patients, increasing the impetus for outpatient CAR T therapy. Our group has over 10 years of experience with clinical trials and commercial use administering CAR T cells in the outpatient setting with the majority of therapy using tisa-cel.^{24, 25} Tisa-cel is particularly conducive to outpatient therapy given the delayed and gradual onset of toxicity allowing early identification of patients who require inpatient care. Here we report on outcomes and safety using commercial tisa-cel primarily administered in an outpatient setting with admission only for toxicity management.

Patients and Methods

We conducted a retrospective study investigating the real-world experience at a single academic medical center of all adult patients receiving commercial CAR T therapy with tisa-cel. Expedited approval of the data collection was provided by the IRB and waiver pf consent was obtained. We examined all patients receiving this modality treatment between 10/2017 and 12/2020 including between 3/2020 and 12/2020 when admission for elective therapy was limited by the SARS-CoV-2 pandemic restrictions. All patients were assessed for eligibility for CAR T therapy based on the FDA label. Many of these patients may not have strictly met criteria for entry into the primary trials for these products due to performance status, age or borderline organ function

Prior to start of therapy, patients are assessed for performance status and volume of disease by the treating physician as well as an internal cellular therapy committee to review eligibility for CAR T. Patients with rapidly progressive disease were reviewed for stability prior to going forward with some patients’ therapy ultimately deferred; others were admitted to the hospital prior to lymphodepleting chemotherapy.

Eligible patients were consented for the product to be used based on disease status and side effect profile. Discussion of location of therapy (inpatient vs. outpatient) was also part of standard procedures made to facilitate outpatient CAR T therapy at our center. Standard procedures included detailed evaluation of performance status and disease status prior to CAR T, greater use of alternative lymphodepletion therapies outside of fludarabine and cyclophosphamide, and close follow up in clinic for the first 30 days (Figure 1). Of note, this analysis was performed before the FDA approval for liso-cel.

Figure 1.

Ambulatory Tisagenlecleucel Practice Pathway. Decision Pathway for Determining Eligibility for Ambulatory CAR T Cell Therapy with Tisagenlecleucel. CAR T = Chimeric Antigen Receptor T-cell.

Indications for initial hospitalization for tisa-cel at our institution included suboptimal organ function at time of tisa-cel infusion, geographic restrictions, lack of family support, or progressive lymphoma symptoms requiring inpatient management. Patients received lymphodepleting therapy as an outpatient, followed by evaluation in clinic and outpatient infusion of tisa-cel. Patients returned regularly after infusion for follow-up on day 2 and day 4, then weekly starting day 8 through day 30 for physical examination, labs, and assessment for cytokine reliease syndrome (CRS) and neurotoxicity. Patients were instructed to contact our clinic with fever > 100.4F, flu-like symptoms such as malaise, or any change in mental status. The attending physician would provide direct contact information for any events occurring outside clinic hours. Patients were also required to stay within 1 hour driving distance of our clinic and to have identified a caregiver who remained with them continuously for the first 28 days. Resources were provided by a nurse coordinator and social worker to facilitate any logistics around the 1-hour radius requirement.

Any concerning call or visit would lead to an in person evaluation at our outpatient clinic or emergency department followed by admission if necessary. CRS presenting as grade 1 was assessed and managed in clinic with higher grade adverse events referred for emergent evaluation and admission. Telemetry and oxygen monitoring were not performed on an ambulatory basis. Our institutional toxicity management protocol for CRS and neurotoxicity were developed to comply with ASTCT guidelines with CRS management using supportive care for grade 1 adverse events with escalation of care for higher grade events.²⁶

Diagnosis codes for CAR T therapy and status of disease at time of infusion were collected as well as the lymphodepletion therapy, location of infusion and number of hospitalizations within 30 days after infusion, and rates of cytokine reliease syndrome (CRS) and neurologic toxicity and other CAR T associated complications. Only patients receiving commercial product at full specification were included. Note was made of any detected COVID infection and any delays or morbidity/mortality related to infection. Data was exported from an electronic database stored in REDCap. Basic statistics and survival analyses were exported through a REDCap dashboard.

Results

There were 72 patients identified for this analysis with 13 additional patient receiving products out of specification who were excluded from our analysis. All subjects had been diagnosed with aggressive lymphoma subtypes approved for CAR T therapy and the median of patients received tisa-cel as their fourth line of therapy, including patients with transformed disease. Baseline characteristics of the population, including median age, NHL subtype, performance status, treatment class used prior to apheresis, bridging status, and bridging treatment class are described in Table 1.

Table 1.

Baseline Patient Characteristics (N = 72)

Patient Characteristics	n (%)
Median age at Tisa-cel infusion, years (Min, Max)	65.7 (32, 83)
Gender
Male	42 (58.3)
Female	30 (41.7)
NHL Subtype
DLBCL	35 (48.6)
Transformed follicular lymphoma (FL)	20 (27.8)
High-grade B cell lymphoma	7 (9.7)
Richter Transformation	3 (4.2)
Transformed unspecified low grade lymphoma	2 (2.8)
Transformed Nodular lymphocyte predominant Hodgkin lymphoma (NLPHL)	1 (1.4)
Primary mediastinal B cell lymphoma (PMBCL)	1 (1.4)
Eastern Cooperative Oncology Group (ECOG) Performance Status
0	23 (31.9)
1	45 (62.5)
2	3 (4.2)
3-4	0
Unknown	1 (1.4)
Lactate Dehydrogenase (LDH) at Time of Tisa-cel Infusion (Reference Range 98-192 U/L)	Value, U/L (Range, U/L)
LDH	183.5 (98-639)
Prior Stem Cell Transplant
Autologous Stem Cell Transplant	13 (18.1)
Allogeneic Stem Cell Transplant	0
Median Line of Therapy of Tisa-cel Infusion
Median Line of therapy, no. (Min, Max)	4 (2, 12)
Median Time from Diagnosis to Tisa-cel Infusion, Years (Min, Max)
Overall	2.5 (0.3, 22.5)
Non-transformed Diagnosis (n = 43)	1.9 (0.3, 18.2)
Transformed Diagnosis (n = 29)	5.2 (0.7, 22.5)
Treatment Class Pre-Apheresis
Cytotoxic Chemotherapy	35 (48.6)
Kinase Inhibitor	13 (18.1)
Monoclonal Antibody	11 (15.3)
Antibody-Drug Conjugate	5 (6.9)
Radiation	5 (6.9)
Other	3 (4.2)
Bridging Therapy
No	17 (23.6)
Yes	55 (76.4)
Bridging Treatment Class
Antibody-drug Conjugate	14 (19.4)
Cytotoxic Chemotherapy	13 (18.1)
Radiation	11 (15.3)
Small Molecule Inhibitor	8 (11.1)
Monoclonal Antibody	4 (5.6)
Immunomodulator + Monoclonal Antibody	4 (5.6)
Radiation + PD-1 inhibitor	1 (1.4)
Lymphodepleting Regimen
Bendamustine	69 (95.8)
Flu/Cy	2 (2.8)
Cyclophosphamide 1000 mg/m2 x 1 d	1 (1.4)
Infusion Setting
Outpatient	68 (94.4)
Inpatient	4 (5.6)

Abbreviations: DLBCL = Diffuse large B-cell lymphoma; NHL = non-Hodgkin lymphoma.

The majority of patients (94.4%) received their lymphodepletion and CAR T infusion as outpatients. The four patients (5.6%) admitted at the time of their T-cell infusion required admission due to their clinical status at the time of infusion, either progressive symptoms due to lymphoma (n = 1) or other comorbidities (n = 3). Of the three patients who received tisa-cel as an inpatient due to comorbidities, two had geographical restrictions and/or no family to monitor them as an outpatient and the last patient had non-neutropenic fevers that required evaluation prior to tisa-cel administration.

Bridging therapy between the collection of T-cells and infusion of the modified cells was administered in 76.4% of patients. Bridging therapy was initiated by the treating physician for reducing tumor burden or palliation of lymphoma symptoms while awaiting CAR T infusion. A range of therapies were used as bridging with no one therapy noted to be superior.²⁷ Response to bridging therapy immediately prior to infusion was not assessed consistently. Formal restaging prior to tisa-cel was not required at our institution and not formally documented. All decisions regarding assessments of disease status prior to therapy was by clinical evaluation only. The ORR response rate for patients requiring bridging was 40% (Table 2) with a median progression-free survival (PFS) of 94 days, no significant difference was found in comparison to those not receiving bridging but a trend was noted for worse outcome (median PFS 114 d vs.94 d).

Table 2.

Efficacy

Best Response to Tisa-Cel	No Bridging (n = 17), n (%)	Bridging (n = 55), n (%)	Overall (n = 72), n (%)
Complete response (CR)	5 (29.4)	20 (36.4)	25 (34.7)
Partial response (PR)	4 (23.5)	2 (3.6)	6 (8.3)
Stable disease (SD)	1 (5.9)	3 (5.5)	4 (5.6)
Progressive disease (PD)	7 (41.2)	30 (54.5)	37 (51.4)
Vital Status at Last Observation
Alive	12 (70.6)	37 (67.3)	49 (68.1)
Deceased	5 (29.4)	18 (32.7)	23 (31.9)

The choice of lymphodepleting chemotherapy was at the discretion of the clinical provider given the FDA label for tisa-cel allows for use of fludarabine/cyclophosphamide, bendamustine, or omission of lymphodepletion for a white blood cell count less than 1 × 10⁹/L within 1 week of reinfusion. As has been reported, bendamustine has lower levels of cytopenias and faster count recovery which may impact symptoms in the post infusion setting.^{28, 29} Complications from CAR T therapy were very manageable with no grade 3-5 CRS reported and majority of CRS toxicity limited to grade 1-2 (40.3%) (Table 3). All therapies for CRS management (eg, tocilizumab and steroids) were administered after evaluation in the emergency department or after direct admission to the oncology unit. 94.4% of patients had no neurologic toxicity and the highest grade of neurotoxicity was grade 3 in two patients (2.8%). Many patients were successfully managed for grade 1 CRS without the need for emergency department visit or admission.

Table 3.

Safety (N = 72)

Cytokine Release	n (%)
None	43 (59.7)
Grade 1	16 (22.2)
Grade 2	13 (18.1)
Grade 3-5	0
Neurotoxicity	n (%)
None	68 (94.4)
Grade 1	2 (2.8)
Grade 2	0
Grade 3	2 (2.8)
Grade 4-5	0
Hospitalization within 30 d of Tisa-cel	n (%)
No	46 (63.9)
Yes	26 (36.1)
Hospitalization within 72 hrs days of outpatient Tisa-cel administration	n (%)
No	58 (90.6)
Yes	14 (19.4)

Only 26 (36.1%) of patients required admission within 30 days of infusion, primarily for evaluation and management of CRS symptoms and suspected infection. Of the patients admitted within 30 days, 22 patients (30.5%) were admitted for CRS (n = 4 grade 1 CRS), 2 patients (2.8%) were admitted for infection, 1 patient (1.4%) was admitted for colitis, and 1 patient (1.4%) was admitted for catatonia found unrelated to tisa-cel infusion. The median length of stay for patients hospitalized within 30 days was 5 days (range 2-23 days). There were 14 patients (19.4%) admitted within 72 hours of infusion. LDH levels at time of infusion of patient’s admitted within 30 days of infusion (184 U/L, range 107-393 U/L) did not differ greatly from the larger population (183.5 U/L, range 98-639 U/L).

During the peak incidence of COVID infections in Philadelphia, PA between 3/2020 and 12/2020, 22 patients successfully received CAR T (tisa-cel). Six patients contracted COVID-19 with two rapid deaths attributed to COVID-19 infection complications. All COVID infections occurred after infusion of CAR cells. Since all patients received a viral workup, including evaluation for SARS-CoV-2 to proceed to cellular infusion, any patients found to have COVID-19 prior to lymphodepletion were delayed until negative. One additional patient had prolonged ongoing hospitalization and ultimately mortality due to the consequences of COVID-19. The remaining patients had recovery from COVID-19 and routine follow up for their lymphoma post CAR T.

After a median follow-up of 9.1 months (range 0.7-29.3 months) in our analysis, the overall response rate for the whole group was 43% with 34.7% achieving CR, comparable to what has been reported previously for tisa-cel.^{18, 19} There was improved outcome if in CR at the time of infusion. For patients in CR as best response, the median PFS was not reached by the time of our analysis (Figure 2A). Median PFS for the entire cohort (Figure 2B) was 3.3 months (range 3.1-9.1 months). The median overall survival stratified by response is depicted in Figure 2C; the median overall survival for the cohort was 26.5 months (range 19 months-not reached) (Figure 2D).

Figure 2.

Progression-Free Survival and Overall Survival for Tisagenlecleucel Therapy. (A) Progression-free survival by best response to tisagenlecleucel. (B) Progression-free survival for study cohort treated with tisagenlecleucel (N = 72). (C) Overall survival by best response to tisagenlecleucel. (D) Overall survival for study cohort treated with tisagenlecleucel (N = 72).

Discussion

CAR T therapy has revolutionized the treatment of patients with relapsed/refractory large cell lymphoma, particularly those who have been chemotherapy resistant. This approach using cellular therapy requires the expertise of experienced centers given the logistics of the preparation for therapy and management of the unique side effect profile including CRS and immune effector cell-associated neurotoxicity syndrome (ICANS). We administer tisa-cel primarily in an outpatient setting with admission only for management of toxicity. In our experience with tisa-cel, we find that only a third of patients required admission within 30 days of infusion, and 19.4% within the first 3 days. It is notable that of the 68 patients treated as outpatients, no patient required treatment of CRS in the emergency department highlighting that patients typically do not have severe CRS on presentation as long as patients are within the recommended treatment radius of infusion center and have adequate social support. Early outpatient management of grade 1 CRS largely obviated the need for evaluation in the emergency department or admission for the majority of patients. While 2 patients died from COVID complications after CAR T cells infusion, no deaths were attributed to outpatient management or delay in care. Close communication with an experienced inpatient cellular therapy team improved the transitions from outpatient administration to an inpatient setting in the setting of acute toxicities.

For ambulatory administration, our center has prioritized the use of 4-1BB costimulatory domain CARs in DLBCL patients due to a more favorable reported toxicity profile and therefore lower chance of a re-admission within 72 hours of cell administration. The reported rate of CRS for tisa-cel in NHL was 58% with 23% experiencing grade 3 or 4 symptoms (Penn Grading system)³⁰ ICANS was reported in 21% with 12% with grade 3 or 4 neurologic adverse events. Median time to onset of CRS with tisa-cel in NHL was a median of 3 days (Range 2-30 days).¹⁹ These were all managed with supportive care.³¹ In addition, CRS, even when more severe, was typically gradual in onset and therefore there was ample time for patients to notify their care team, and seek medical evaluation.

Our outcomes demonstrate that patients can receive CAR T with tisa-cel effectively and safely as an outpatient, in the context of close clinical supervision. Training in the REMS program is required for all team members. Best practices and standard operating procedures are regularly reviewed and quality measures are emphasized. As mentioned in previously reported best practice recommendations,³² a multidisciplinary approach has been in place at our center to best manage patients in an ambulatory setting, including: coordinators, administrative assistants, nurses, advanced practice providers, pharmacists, and physicians. Careful attention should be paid to the performance status of the patient and the status of disease prior to lymphodepletion and T cell infusion.

From the onset of the COVID-19 pandemic, the rate of consultation and level of care at our center remained unchanged from a prepandemic setting. Given limitations on bed availability in the hospital and particularly in the ICU, outpatient administration allowed for continued care of this refractory population. Our team worked to monitor SARS-COV2 exposure closely during the CAR T process with strict attention to symptoms and viral testing.

In conclusion, outcomes for patients treated with tisa-cel at our center using this approach approximate previously reported outcomes for subjects receiving CAR T on study.^{18, 19, 33-35} Shift of therapy to the outpatient setting for these patients may expand the feasibility of these therapies and potentially reduce overall costs and resource use. In the setting of limited hospital beds, as experienced during the COVID 19 pandemic in 2020 and 2021, outpatient CAR T therapy also reduces the burden on the health care system while still permitting aggressive and potentially curative therapy.

Clinical Practice Points.

• Cellular therapy with CD-19 directed CAR T-cells transformed management of relapsed/refractory DLBCL.

• We present our institutional approach to management of tisa-cel in an ambulatory setting of care outside the setting of a clinical trial.

• Outcomes in our population are similar to the registrational study for tisa-cel and subsequent real-world analyses.

• Hospital admissions for acute toxicity in the first 30 days of our study were low (36.1%) and grade 3-4 toxicities, including CRS and neurotoxicity were uncommon.

• Ambulatory administration of tisa-cel is both safe and effective and can help alleviate inpatient resources, especially in the setting of the SARS-CoV2 pandemic.