Outpatient CD19‐directed CAR T‐cell therapy is feasible in patients of all ages

The emergence of CAR T-cell therapy (CAR-T) has revolutionized the treatment paradigm for B-cell malignancies.1, 2 Due to concern for toxicity, most CAR-T applications thus far have taken place in a tightly supervised inpatient setting. This is especially true in older patients, who often have a higher comorbidity status and may need closer monitoring. Outpatient administration of CAR-T, however, offers the opportunity for greater patient autonomy and lower healthcare-related costs. Hence, in this study, we evaluate our institutional experience with the use of CAR-T in an outpatient setting for all age groups, demonstrating the feasibility of outpatient CAR-T even in older patients. Our study included all patients who received commercial CAR-T in an outpatient setting at Johns Hopkins University between January 2019 and July 2022 with a data cut-off date of 31 August 2022. All patients irrespective of CAR-T product are planned for outpatient infusion at our centre unless clinical status warrants inpatient stay at the time of planned infusion. Per institutional guidelines, patients receive lymphodepletion (LD) chemotherapy and CAR-T infusion in an outpatient setting followed by daily monitoring in our extended-hours outpatient unit for 14 days (Figure S1). This unit is monitored by a 24-h triage service that can directly admit patients to the inpatient unit as needed for toxicity management. Overall response rate (ORR) included complete response (CR) or partial response. Response grading followed standard criteria.3, 4 Cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS) were graded using Consensus Grading criteria.5 Recovery for blood counts was defined as grade 1 leukopenia without growth factor support in 2 weeks, grade ≤2 anaemia without red cell transfusion in 2 weeks, grade ≤2 thrombocytopenia without transfusions in 1 week, and grade ≤2 neutropenia using Common Terminology Criteria for Adverse Events, Version 5.0.6, 7 Age cohorts were compared using Fisher's exact test, Wilcoxon rank-sum test and Chi-square test as appropriate. Overall survival (OS, event: death) and progression-free survival (PFS, events: progression and death) after CAR-T were estimated based on the Kaplan–Meier method and compared between age cohorts by log-rank tests. During the study period, 52 patients received CAR-T at our centre. Five patients received inpatient LD, CAR-T and postinfusion care for symptomatic disease progression and were excluded from this analysis. Of the 47 patients included in this study, 14 (30%) were ≥65 years old (yo) and 33 (70%) <65 yo. Twelve (86%) patients ≥65 yo and 25 (76%) patients <65 yo received axicabtagene ciloleucel (axi-cel) or brexucabtagene autoleucel (brexu-cel). To assess the safety profile of axi-cel prior to instituting the full outpatient treatment paradigm, the first seven patients who received this product at our centre were admitted pre-emptively post-CAR-T infusion to facilitate closer monitoring. Another patient with axi-cel had symptomatic disease progression and was therefore pre-emptively admitted after receiving CAR-T infusion for clinical monitoring. Table 1 provides baseline characteristics and treatment outcomes. Median follow-up was 364 days (range 37–1324). Hospitalization rates, including pre-emptively admitted patients, were 12/14 (86%) in patients ≥65 yo and 27/33 (82%) in patients <65 yo (p > 0.99). Hospitalized patients ≥65 yo stayed a median of 10 days (range 2–29 days) while those <65 yo stayed a median of 7 days (range 3–33 days) (p = 0.88). The median time to hospitalization for patients ≥65 and <65 yo were 4 days (range: 0–14) and 2 days (range: 0–10 days) respectively (p = 0.15). No patients ≥65 yo were admitted to the ICU, while 4/27 (15%) of the hospitalized patients <65 yo required ICU admission. Toxicity outcomes by age and severity are shown in Figure 1. CRS occurred in 10/14 (71%) of patients ≥65 yo versus 25/33 (76%) of patients <65 yo (p = 0.73). All were grade 1–2 except one <65 yo patient, who was grade 3. Median duration of CRS was 4.5 days (range 1–7 days) for patients ≥65 yo versus 4 days (range 1–11 days) for patients <65 yo (p = 0.46). Tocilizumab was given in 8/10 (80%) of patients ≥65 yo versus 17/25 (68%) of patients <65 yo (p = 0.69). ICANS occurred in 7/14 (50%) of patients ≥65 yo versus 9 (27%) of patients <65 yo (p = 0.24); grade ≥3 ICANS was reported in 3/14 (21%) of patients ≥65 yo and 3/33 (9%) of patients <65 yo (p = 0.34). Median duration of ICANS in patients ≥65 yo was 10 days (range 3–25) while it was 5 days (range 1–16) in patients <65 (p = 0.13). With axi-cel, rates of all-grade and grade 3–4 ICANS were 8/14 (57%) and 4/14 (29%) in patients ≥65 yo, and 6/22 (27%) and 3/22 (14%) in patients <65 yo respectively. One patient ≥65 yo (axi-cel) and two patients <65 yo (brexu-cel) developed immune effector cell-associated haemophagocytic lymphohistiocytosis-like syndrome.8 WBC, ANC and haemoglobin recovery after 30 days showed no significant age group differences. All patients ≥65 yo achieved 4-week platelet recovery versus 76% in patients <65 yo (p = 0.08). Patients ≥65 yo had a 30-day ORR of 7/14 (50%) and patients <65 yo had a 30-day ORR of 15/33 (45%). Additional clinical outcomes are elaborated in Table 1. OS and PFS are depicted in Figure S2. Our experience, while limited by sample size, underscores the feasibility of outpatient CAR-T for all patients, regardless of age, with appropriate infrastructure for close monitoring and direct admission, if needed. Despite an 83% admission rate, the median duration of admission was limited to a median of 7 and 10 days for patients <65 and ≥65 yo, which is shorter than that previously reported for inpatient CAR-T administration.9 These data substantiate the growing interest in outpatient CAR-T owing to its favourable financial sustainability and healthcare resource utilization profile.10 Notably, more frequent higher grade and protracted course of ICANS can be seen in older patients as noted in our data and in other series.11-13 Hence, adequate resource allocation for close inpatient monitoring when needed is critical. Compared to ZUMA-1, we found overall lower rates of severe CRS and ICANS.14 This is likely a result of evolution of real-world practice and incorporation of early intervention, which may make outpatient CAR-T more feasible. Interestingly, our response rates appear lower than other studies, possibly resulting from our report of 30-day response rather than best response as reported in other studies.14, 15 Additionally, our institutional practice has been to consider allogeneic blood or marrow transplantation in earlier lines (23% in our series compared to less than 5% in most other series),16 suggesting that patients likely had more aggressive disease at the time of CAR-T. Prophylactic steroids and future investigations of additional prophylactic strategies for CRS and ICANS may enable greater use of outpatient CAR-T over time.17 This research received no external funding. Andrew Ly and Khaled Sanber collected the data, interpreted the data and wrote the manuscript. Andrew Ly, Khaled Sanber and Hua-Ling Tsai conducted statistical analyses. Angela Ondo, Kathy Mooney, Audra Shedeck, Philip Hollingsworth Imus, Julie Baker and Nina Wagner-Johnston interpreted the data and reviewed the manuscript. Tania Jain conceptualized the idea, interpreted the data and wrote the manuscript. T.J. receives institutional research support from CTI Biopharma, SyneosHealth and Incyte and participates in advisory boards with Care Dx, Bristol Myers Squibb, Incyte, Abbvie, CTI, Kite, Cogent Biosciences, Blueprint Medicine and Protagonist Therapeutics. N.W.J. has participated in advisory boards with Beigene and receives research funding from Genentech, Merck, AstraZeneca, ASTEX and ADC Therapeutics. Approval for this study was granted by the Johns Hopkins Medicine Institutional Review Board (IRB00325578). The data that support the findings of this study are available from the corresponding author upon reasonable request. Figure S1. Figure S2. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.