The Impact of Outpatient versus Inpatient Administration of CAR-T Therapies on Clinical, Economic, and Humanistic Outcomes in Patients with Hematological Cancer: A Systematic Literature Review

Simple Summary The administration of chimeric antigen receptor (CAR)-T cell therapies in an outpatient setting is expanding; however, there is limited evidence comparing outcomes from outpatient and inpatient administration. This research aims to compare the clinical and economic outcomes associated with the administration of CAR-T therapies in either setting for patients with hematological cancer, by summarizing existing evidence, and also to test the hypothesis that outpatient administration of CAR-T therapies yields comparable clinical effectiveness as inpatient treatment while offering significant cost reduction and lower humanistic burden. Abstract Although chimeric antigen receptor (CAR)-T cell therapies are typically administered in the inpatient setting, outpatient administration is rapidly expanding. However, there is limited summarized evidence comparing outcomes between outpatient and inpatient administration. This systematic literature review aims to compare the safety, efficacy, quality of life (QoL), costs, and healthcare resource utilization (HCRU) outcomes in patients with hematological cancer who are administered CAR-T therapy in an outpatient versus an inpatient setting. Publications (2016 or later) that reported the outcomes of interest in patients treated with a CAR-T therapy in both outpatient and inpatient settings, or only the outpatient setting, were reviewed. In total, 38 publications based on 21 studies were included. Safety findings suggested the comparable frequency of adverse events in the two settings. Eleven studies that reported data in both settings showed comparable response rates (80–82% in outpatient and 72–80% in inpatient). Improvements in the QoL were observed in both settings while costs associated with CAR-T therapy were lower in the outpatient setting. Although unplanned hospitalizations were higher in the outpatient cohort, overall HCRU was lower. Outpatient administration of CAR-T therapy appears to have comparable outcomes in safety, efficacy, and QoL to inpatient administration while reducing the economic burden.


Introduction
Hematological cancers are a significant global health issue characterized by high mortality rates [1,2].They negatively affect patients' life expectancy and quality of life (QoL) and impose a substantial economic burden [3][4][5][6].The clinical outcomes for hematological

Data Analysis
Evidence identified from the systematic literature search was analyzed qualitatively.The compiled evidence was tabulated, summarized, and presented graphically for the following elements of the research: study details (trial design, tumor type, treatment setting, sample size, and follow-up duration) and outcomes presented (safety, efficacy, QoL, HCRU, and costs incurred).Safety outcomes included the cytokine release syndrome (CRS), neurologic toxicities, and other toxicities reported in individual publications while the collated efficacy outcomes included the complete response (CR), partial response (PR), overall response rate (ORR), progression-free survival (PFS), and overall survival (OS).HCRU measures reported included the rate of, time to, and reasons for hospitalization, length of hospital stay, rate of ICU admissions and length of stay, and outpatient visits.Costs incurred in different follow-up periods post-infusion were compiled and categorized as available.
The data were reported separately for outpatient and inpatient cohorts and included evidence from comparative as well as single cohort studies.

Literature Search Results
Database searches identified 7701 initial records.After deduplication, 5648 records remained for screening against inclusion and exclusion criteria.A total of 1125 records met the relevant criteria and an additional 83 records were obtained from supplementary sources including Google Scholar, conference proceedings, and a bibliography of identified studies and SLRs, resulting in a total of 1208 records.Ultimately, 38 records that reported outcomes for patients who underwent infusion/management in the outpatient setting or in both outpatient and inpatient settings were considered for qualitative synthesis (Figure 1).sources including Google Scholar, conference proceedings, and a bibliography of fied studies and SLRs, resulting in a total of 1208 records.Ultimately, 38 records t ported outcomes for patients who underwent infusion/management in the outpatie ting or in both outpatient and inpatient settings were considered for qualitative syn (Figure 1).

Study Characteristics
The 38 included records were based on 21 unique studies, most of which wer lished in 2022 and 2023.The patient populations of these studies included indiv diagnosed with ALL, various types of lymphoma, such as B-cell lymphoma (BC follicular lymphoma (FL), and multiple myeloma.

Study Characteristics
The 38 included records were based on 21 unique studies, most of which were published in 2022 and 2023.The patient populations of these studies included individuals diagnosed with ALL, various types of lymphoma, such as B-cell lymphoma (BCL) and follicular lymphoma (FL), and multiple myeloma.

Quality Assessment
All of the 21 studies included in the analysis underwent a quality assessment using relevant checklists based on their study design.The assessment aimed to evaluate the risk of bias and the overall quality of the studies.The ROB 2 checklist was used to assess the RCTs conducted by Kamdar in 2022 [64].The findings from this assessment revealed a high risk of bias in the trial (Table S4).The modified Downs and Black checklist with 27 items assessed eight non-randomized single-arm trials.Table S5 highlights one study of good quality (score, [15][16][17], five of fair quality (score, [12][13][14], and two of poor quality (score < 11) [38,39,42,46,49,51,67].The NOS criteria were used to assess the quality of double-arm and single-arm observational studies.Of the six double-arm studies, five were considered of good quality (score > 6) and one of fair quality (score 5) (Table S6).Among the six single-arm studies, five were rated as good quality (score > 4) and one as fair quality (score 4) [26,41,45,[53][54][55]57,59,60,62,63,68].

Safety
The safety of CAR-T therapy was evaluated in several studies, with data available for both outpatient and inpatient settings [37,39,42,53,67].In total, 5 publications provided safety data for both settings while 10 publications reported data solely on patients treated in the outpatient setting.In the publications reporting data for both settings, the frequency of AEs was generally comparable or higher for patients who received CAR-T in the inpatient setting.Two publications by Denlinger (2022) [53] and Sehgal (2022) [39] reported a higher frequency of CRS in patients who received CAR-T in the inpatient setting whereas the other three publications found similar rates of CRS between the two cohorts.The OUTREACH and ELARA trials reported other AEs, including infections, leukopenia, anemia, and thrombocytopenia.In the OUTREACH trial, the incidence of infections was comparable between the outpatient and inpatient settings, with 33% and 32% of patients experiencing infections, respectively.Additionally, prolonged cytopenia at the Day 29 visit was also consistent between outpatient and inpatient settings, with 33% and 32% of patients affected, respectively.In the ELARA trial, the incidence of infections was lower in the outpatient setting compared with the inpatient setting, with any grade infections reported in 17.6% of outpatient versus 21.3% of inpatient cohorts.Although none of the patients in the outpatient group experienced a Grade 3-4 infection, 7.5% of those in the inpatient group experienced such severe infections.Furthermore, the ELARA trial indicated a higher occurrence of hematological disorders in the inpatient setting, with any grade cytopenia reported in 77.5% of inpatient compared with 64.7% of outpatient administration groups.Similarly, the occurrence of Grade 3-4 cytopenia was lower in the outpatient group (47.1%) versus the inpatient group (73.8%).Three publications reported a higher frequency of neurologic toxicities among patients who received CAR-T in the inpatient setting [39,53,67].Conversely, one publication by Abramson (2020) [42] reported a substantially higher frequency of neurologic toxicities among patients who received CAR-T in the outpatient setting (Table 3).Most of the publications pertaining to only the outpatient cohort reported high frequencies of CRS-related (40-92%) and neurologic (6-57%) toxicities [45,[47][48][49]51,55,57,59,60,63].Detailed safety outcomes are presented in Table 4.For studies involving CAR-T treatment in both inpatient and outpatient settings, three publications reported no deaths from CRS and neurological toxicities [37,39,42].For studies based on treatment in only the outpatient setting, Turtle et al. reported a 5% mortality rate, with most deaths occurring during the dose-finding phase, primarily due to CRS and neurological events [51].In the studies reporting response outcomes from both settings, the overall response rate (ORR) was found to be comparable between patients managed in the outpatient setting (ORR, 80-82%) and those managed in the inpatient setting (ORR, 72-80%) [36,37,39,42].In the OUTREACH trial, the ORR was 82% in the outpatient cohort and 76% in the inpatient cohort [36,37].In the PILOT trial, the ORR was 80% in both outpatient and inpatient groups [39].In the TRANSCEND NHL 001 trial, the ORR was 80% in the outpatient cohort and 72% in the inpatient cohort [42].The response rates observed in real-world studies (43-88%) were lower than those observed in clinical trials (80-100%) for patients managed in the outpatient setting.The median duration of response was found to range from 8.2 to 15.1 months among the CAR-T-treated patients managed in the outpatient setting and were comparable to the duration of response in those managed in the inpatient setting (12.1 to 14.8 months) (Table 5) [36,37,39].Progression-free survival (PFS) data for patients managed in the outpatient setting were reported in a limited number of studies.In the OUTREACH trial, patients receiving treatment in the outpatient setting had a 12-month PFS rate of 41% which was similar to the rate observed in an inpatient setting (39%).In the ELARA trial, the 12-month PFS rate was 60% for patients in the outpatient group compared with 70% for patients receiving treatment in an inpatient setting (Table 6) [37,39,42,63,67].Similarly, the overall survival (OS) data for patients managed in the outpatient setting were limited.The Kaplan-Meier estimates for 12-month OS ranged from 60% to 75% among patients with lymphoma managed in the outpatient setting and were reported to be 60% among those managed in the inpatient setting [37,39,47,60,63].Notably, a real-world study reported a median OS of 26.5 months in patients with lymphoma managed in the outpatient setting (Table 7) [60].2022) reported QoL assessments in patients treated with CAR-T therapy at multiple time points, including pre-treatment (baseline), on day of infusion, on several dates after infusion (days 29, 50, 90, 180, 270, 365, 545, and 730/end of study), and at disease progression and assessed the least squares (LS) mean change from the baseline for visits with ≥10 patients.Both the outpatient and inpatient groups exhibited comparable meaningful improvements in various aspects of QoL, including global health status/QoL, fatigue, pain, assessed by EORTC-QLQ-C30, and the EuroQol 5-Level 5-Dimension questionnaire (EQ-5D-5L) visual analog scale (VAS) scores (Table 8) [38].

Direct Costs
Seven studies reported costs/reimbursement amounts associated with CAR-T administration in both outpatient and inpatient settings in the United States (Table 9) [26,41,52,53,62,65,67].Of these, six reported that the post-infusion costs were lower for patients who received CAR-T in an outpatient setting compared with those who received it in an inpatient setting [41,52,53,62,65,67].Costs in the 6-month post-infusion period reported from TRANSCEND NHL 001, OUTREACH, TRANSFORM, and PILOT trials revealed to be two to four times greater costs in the inpatient setting-these ranged from USD 61,772 to USD 96,297.Pooled analysis of TRANSCEND NHL 001 and OUTREACH trials similarly showed a lower 6-month post-infusion cost in the outpatient setting (USD 36,702) compared with the inpatient setting (USD 89,535) [52].The main driver for the higher costs in the inpatient setting was identified to be hospitalization costs [41,52,65].Yang et al. (2022), the only study that assessed the Medicare reimbursement amount, including CAR-T product costs, reported that costs in the first month were nominally higher in the outpatient cohort compared with the inpatient cohort and were comparable in subsequent months [26].* Denlinger et al. did not specify the settings; however, it seems that tisa-cel was administered in the outpatient setting and axi-cel in the inpatient setting.** Reimbursement amounts were assessed only among patients who received CAR-T in non-prospective payment system-exempt hospitals.Abbreviations: 2L+, second line and greater; 3L+, third line and greater; axi-cel, axicabtagene ciloleucel; BCL, B-cell lymphoma; CAR-T, chimeric antigen receptor T-cell; DLBCL, diffuse large B-cell lymphoma; ER, emergency room; FL, follicular lymphoma; FU, follow-up; hosp., hospitalization; ICU, intensive care unit; IP, inpatient; LBCL, large B-cell lymphoma; LOS, length of stay; NA, not available; OP, outpatient; Ph, phase; RWE, real-world evidence; R/R, relapsed/refractory; tisa-cel, tisagenlecleucel; US, United States.

Healthcare Resource Utilization
The analysis of HCRU among patients who received CAR-T in the outpatient setting was reported in 17 publications [26,37,39,41,46,[52][53][54][55]57,59,60,[62][63][64]67,68].Of these, 10 also reported data for patients who received CAR-T in the inpatient setting (Table 10) [26,37,39,41,[52][53][54]62,67,68].Most of the HCRU data were related to inpatient admissions, length of stay (LOS), and intensive care unit (ICU) admissions.Some publications also reported a median time to hospitalization for the outpatient cohort, ranging from 4 to 9 days [37,39,52,55,57,64,67].Chihara et al. (2022) and Wright et al. (2020) reported lower rates of unplanned hospitalization (defined as post-CAR-T infusion in the outpatient cohort and re-hospitalizations in the inpatient cohort) in the inpatient cohort [54,68].Despite such higher unplanned hospitalization observed in the outpatient cohort, the overall HCRU was lower for the outpatient cohort in publications that reported data for both cohorts.Most studies reported higher overall mean LOS for the inpatient cohort, with two to three times longer durations compared with the outpatient cohort across studies [26,37,39,41,52,53,62,67]. Additionally, ICU utilization was found to be higher for the inpatient vs. outpatient cohort, as shown by Fowler et [26,41,52,53,62,65,67].Similarly, the mean ICU LOS was two to three times higher for the inpatient cohort across studies, except the study by Sehgal et al. (2022) which reported a lower ratio (0.8) of ICU LOS for the inpatient cohort [39].Additional HCRU data available in the identified studies are presented in Table S7.

Discussion
This comprehensive SLR on CAR-T therapies in patients with hematological cancer highlights the potential benefits of outpatient compared with inpatient administration.Safety outcomes, a key consideration in CAR-T treatments and a principal driver for traditionally treating patients in the inpatient setting, were comparable between those treated in the outpatient setting and those treated in the inpatient setting.Moreover, the analysis revealed comparable effectiveness outcomes between the two settings, including the response rates, duration of response, and survival outcomes where reported.Furthermore, both outpatient and inpatient cohorts experienced meaningful improvements in QoL measures.These findings collectively provide compelling evidence to clinicians and other decisionmakers to actively consider administering CAR-T therapies in the outpatient setting for patients whose disease and clinical characteristics permit this.Treating institutions may upgrade their processes and protocols to encourage treatment in the outpatient setting, including preparing for and managing early complications.
To further bolster the case for outpatient administration of CAR-T therapies, assessment of such administration in the real world with longer-term follow-up to allow for the evaluation of survival outcomes such as PFS and OS is warranted.These endpoints are crucial in evaluating the long-term benefits and potential risks associated with the two treatment settings [69].
The observed enhancement in QoL reported by Linhares et al. (2022) likely represents the comprehensive impact of CAR-T therapy on hematological cancer patients, regardless of the treatment setting [38].Further research is needed to explore the underlying factors contributing to the improvement in QoL, paying particular attention to factors related to reduced hospitalization as this could enhance the overall patient experience and improve their QoL during treatment.It is also important to evaluate QoL at multiple time points after CAR-T therapy use and examine if the resolution of AEs reflects improved QoL outcomes.Furthermore, the utilization of other measures, such as the hospital anxiety and depression scale depression subscale (HADS-D), in patients with hematological cancer is important in understanding and addressing their mental health needs more specifically [70].Additionally, a previous SLR emphasized the importance of identifying patients' preferences for involvement in cancer treatment decisions.Establishing these preferences will encourage the healthcare system to become more responsive to individual patient needs and expectations and ultimately, contribute to improving their QoL [71].
The economic outcomes indicate that outpatient treatment may offer cost advantages over inpatient treatment, with patients treated in the latter incurring two to four times higher costs (USD 62,000-96,000) than by outpatient-treated patients (USD 16,000-38,000) [41,52,65].The lower costs observed in the outpatient cohort were primarily driven by reduced hospitalization costs [41,62,65,67].These findings align with an economic evaluation of CAR-T therapy based on the site of care, wherein outpatient CAR-T administration resulted in a substantial decrease in total costs (by 40.4%), with notable reductions observed in hospitalization, office visits, and procedural expenses [72].However, a study by Yang (2022) found that the Medicare reimbursement in the outpatient cohort was slightly higher than that in the inpatient cohort during the first month post-infusion and comparable in subsequent months.The authors, however, attributed this to the inadequate reimbursement for CAR-T infusion in the inpatient setting whereas the outpatient setting reimbursement, which is covered under Medicare Part B, covers not only the CAR-T product cost more completely but also the handling, storage, and a portion of the physician's service fees.Additional efforts are recommended to improve the reimbursement structure and care policies for CAR-T therapy in either infusion setting to suitably incentivize providers [26].
The outpatient cohort experienced more unplanned hospitalizations, with CRS being the main reason for hospitalization.However, the overall HCRU was lower in cases where data were available for both settings as the inpatient cohort exhibited longer stays and a higher HCRU [26,37,39,41,46,[52][53][54][55][56][57]59,60,[62][63][64]67,68].An improved understanding of the predictive risk factors for CRS and neurotoxicity development, including patient disease characteristics and clinical status, can influence personalized decisions regarding outpatient administration as it may offer potential benefits in terms of overall resource utilization.
Outpatient CAR-T administration can potentially expand treatment access by freeing up inpatient capacity and addressing geographic obstacles.A prior economic model concluded that lower costs through outpatient administration could enable more patients to receive treatment with limited resources.[72].
This review provides a comprehensive analysis of both clinical and economic outcomes derived from clinical trials and observational studies concerning CAR-T therapies in a broad patient population with hematological cancer.The review was conducted in accordance with a predefined protocol, with clear inclusion and exclusion criteria, and adhered to the Cochrane guidelines for systematic review reporting.A comprehensive search strategy was employed to minimize reporting bias in the review process.
While this SLR adhered to rigorous selection criteria, it had some limitations.The included studies exhibited heterogeneity in terms of methodology and populations, which prevented direct comparisons.Additionally, in studies that reported outcomes related to the two settings, patients were not randomized between the settings, introducing the potential for bias and raising concerns about the comparability of reported outcomes.In line with the objective of this study, outcomes data were presented here only if reported by setting.The molecular aspect of the therapy has not been discussed as none of the identified publications referred to it as either the driver for the decision of inpatient vs. outpatient administration or as the cause of any difference in the outcomes.Furthermore, patient characteristics were not available for all the studies as the majority of the identified publications were conference abstracts with minimal information on patient characteristics.

Conclusions
Findings from our study showed comparable overall outcomes in safety, efficacy, and QoL between outpatient and inpatient CAR-T administration.While CAR-Ts are typically administered in an inpatient setting, outpatient administration of CAR-T can provide a reduced economic burden without negatively impacting clinical outcomes and should be actively considered where patient disease characteristics and logistical considerations permit this.Future research is needed to explore the impact of administration settings of new CAR-T therapies on patients with multiple myeloma and other hematological cancers.

Supplementary Materials:
The following supporting information can be downloaded at https: //www.mdpi.com/article/10.3390/cancers15245746/s1.Table S1.MEDLINE search strategy; Table S2.Embase search strategy; Table S3.Cochrane search strategy; Table S4.Quality assessment of randomized controlled trial using the Risk-of-Bias version 2 checklist; Table S5.Quality assessment of non-randomized clinical trials (single-arm) using Downs and Black checklist; Table S6.Quality assessment of observational studies (single-arm and double-arm) using the NOS checklist; Table S7.Additional healthcare resource utilization data available in identified studies.

Figure 1 .
Figure 1.The preferred reporting items for systematic reviews and meta-analysis (PRISM diagram.

Figure 1 .
Figure 1.The preferred reporting items for systematic reviews and meta-analysis (PRISMA) flow diagram.

Table 1 .
Characteristics of included studies.

Table 3 .
Safety data in inpatient and outpatient settings reported in the included studies.

Table 4 .
Safety data in outpatient settings reported in the included studies.

Table 5 .
Detailed response results from the studies identified.

Table 6 .
Progression-free survival data reported in the included studies.

Table 7 .
Overall survival data reported in the included studies.

Table 8 .
Quality of life in the OUTREACH trial.
Higher scores on GH/QoL and EQ-5D VAS reflect better QoL.Higher scores on the functioning scales (physical, role, and cognitive) indicate better QoL.Higher scores on the symptom scales (fatigue and pain) indicate worse QoL.Abbreviations: CI, confidence interval; EORTC QLQ-C30, European Organization for Research and Treatment of Cancer Quality of Life Questionnaire-Core 30; EQ-5D-5L, EuroQol 5-level 5-Dimension Questionnaire; GH, Global Health; HUI, Health Utility Index; LS, least-squares; QoL, quality of life; VAS, visual analog scale.

Table 9 .
Costs associated with patients receiving chimeric antigen receptor T-cell therapy in outpatient and inpatient settings.

Table 10 .
Healthcare resource utilization associated with patients receiving chimeric antigen receptor T-cell therapy in the outpatient and inpatient settings.