Chimeric antigen receptor (CAR) T cells are an innovative and promising new class of treatments for patients with hematologic malignancies. This approach has already replaced standard of care (SOC) in early large B-cell recurrence, and continues to challenge immunochemotherapy, stem cell transplantation and even bispecific antibodies. This rapidly evolving landscape, in addition to health-economic considerations, raises the question of how to further evaluate these innovations.

Beyond assessing efficacy and tolerability, other aspects of health, including the patient’s experience, are now essential to appreciate the full impact of these therapies within a one-health perspective. Patient-reported outcome measures (PROMs), such as health-related quality of life (HRQoL), have become increasingly important for health authorities and clinicians in differentiating among treatment options [1]. Despite providing a better understanding of the patient's global status during treatment [2,3], there is surprisingly little HRQoL data in trials assessing CAR T cells in hematologic malignancy. Furthermore, the lack of a gold standard for assessing HRQoL leads to inconsistent methodology and assessment [4,5].

This study aimed to assess the role of HRQoL as an endpoint in trials registered on the US National Institutes of Health (NIH) ClinicalTrials.gov, evaluating the efficacy and safety of CAR T-cell therapy for hematologic malignancies. The results provide a basis for improving the design of future studies to better meet clinicians’ and health decision-makers’ need for knowledge.

The registry search identified 1281 records for hematological malignancies. In total 429 relevant clinical trials were finally selected: 121 in lymphoma, 97 in leukemia, 80 in multiple myeloma, and 131 in two or more of these hematological diseases.

The category "two or more hematological malignancies" mainly concerned leukemia research trials.

This study is the first to report the use of HRQoL in clinical trials assessing CAR T cells in patients with hematologic malignancy registered in the US National Institute of Health's clinicaltrials.gov, with multiple endpoints analyzed.

CAR T cells have achieved unprecedented speed of access, after two phase-2 trials demonstrating their efficacy as third-line or more in patients with DLBCL [15,16] or ALL [17]. Their high cost per infusion [18] and health-economic impact, their positioning in increasingly early lines of treatment for more indications, and their challenging position in relation to both SOC and bispecific antibodies require strengthened evaluations, notably by introducing new endpoints such as HRQoL, and promoting real-life studies to supplement the results of clinical trials [19].

The present study highlights the dramatic inequality in access to CAR T cells worldwide, with >80% of trials conducted in the US and China. Only 11% of studies consider HRQoL as a secondary outcome, although this rate is higher than the 6.8% previously reported [20]. Documenting HRQoL and improving it through early supportive care could prevent treatment discontinuation [21]. Furthermore, this study is the first to reveal that the countries most actively developing CAR T-cell trials are those least interested in quality of life.

The present search found a wide variety of scales used to measure HRQoL, with a preponderance of the EORTC QLQ-C30 [6] and EQ-5D-5Lb [7]. The NIH PROMIS [8–22], which collects PROMs, and SF-36 [9] are also used to assess health status independently of pathology and treatment. The cancer-specific scales used in the clinicaltrials.gov trials are also consistent with the literature: Fact-Lym G [10] in lymphoma, QLQ-MY20 [12] in multiple myeloma, and PedsQL [11] for pediatric populations. Some authors use two scales for HRQoL measurement [4,5,23], although this raises issues regarding the burden imposed by the volume of data for patients and caregivers. The recent MDASI-CAR [24] scale, specific to CAR T cells, could be an interesting alternative; however, it does not allow comparison of CAR T cells versus other treatments such as SOC, stem-cell autograft, or bispecific antibodies.

Data are lacking on the duration of HRQoL collection and the time points. Duration of collection varied widely, from 2 to 15 years, with time points at baseline, month 1, month 3 and then every three months in most cases. In previous studies, data collection duration ranged from 90 days after treatment to five years [4,22,23,25]. The baseline is not standardized: prior to conditioning or salvage chemotherapy [4], patient eligibility screening [26], apheresis and/or before lymphodepletion [22], or renewed on the day of reinfusion [23]. Data collection timepoints included monthly assessments for the first 3 months post-infusion followed by quarterly evaluations [23], monthly assessments for 6 months followed by quarterly evaluations up to 24 months [24], or quarterly assessments up to 12 or 18 months post-infusion [26].

Various methods were used to assess HRQoL: self-administered paper questionnaires [23,27] by patients with or without the help of a caregiver (proxy) [23] or digital applications [5,28]; in some cases, methods were not specified [4]. Self-administered paper questionnaires incur a risk of data loss, non-completion and error when the data are re-entered electronically. Electronic collection seemed to be the most popular, encouraging the use of digital PROMs [28] but is less cost-accessible [27].

However, using an appropriate methodology, the results are highly significant, as shown by the pivotal ZUMA-7 [4], JULIET [26], and TRANSCEND [23] trials, with compliance rates of 80% and 90% in the multiple myeloma CARTITUDE-1 [29] and KARMMA [28] trials, respectively.

Compliance rates were lower in case of disease progression and decreased significantly with the length of follow-up [26].

HRQoL is essential for personalized management and can help clinicians choose between therapeutic strategies. In the pivotal randomized phase-3 ZUMA-7 study [4], axi-cel resulted in significant and clinically relevant improvement in HRQoL compared to SOC at days 100 and 150, faster recovery (by day 150 versus month 9), and less absenteeism and activity impairment at day 100, suggesting a faster return to baseline functioning in the workplace and at home. In the phase-I TRANSCEND [23] trial assessing lisocabtagene maraleucel (lisocel), an initial decline in HRQoL was observed during the first month for six months, followed by stabilization. In the phase-3 TRANSFORM study comparing lisocel versus SOC [5], patients in the lisocel arm showed clinically significantly better overall health and HRQoL.

HRQoL allows us to understand the experience of patients and to compare symptom burden and cognition reported after CAR T-cell treatment versus other strategies [30], and to investigate associations between perceived prognosis, quality of life and mood [31].

These findings call for consensus on HRQoL assessment methodology to increase the use of HRQoL as an endpoint in clinical trials and real-world studies. Guidelines are already available and can inform future developments. Chakraborty et al. provided recommendations for HRQoL collection time points tailored to the CAR T patient's care pathway: at baseline prior to lymphodepleting chemotherapy, at least weekly during the acute phase after CAR T-cell infusion, monthly until one year, and yearly thereafter [19]. Clinicians and scientific societies need to use these initial publications to harmonize methods for collecting and analyzing PROMs and HRQoL. Consensus is now crucial on the choice of HRQoL scale, collection duration and time points, definition of baseline, format, and the number of questionnaires to be completed by patients. The ethical question of whether to collect HRQoL data in patients whose condition is deteriorating will need to be clarified. HRQoL analysis should be integrated as standard practice in clinical trials and real-world studies. Finally, studies should be developed worldwide to promote equitable access to innovation.

This study is based on the ClinicalTrials.gov database. As HRQoL is not a primary endpoint in trials, there is a potential risk of incomplete data collection. Extracting and analysing the data with two independent reviewers should mitigate this risk. The database was last consulted on August 28th, 2022, and the consideration of quality of life in CAR T-cell trials may have evolved since then. It would be interesting to update this work to verify whether HRQoL is being considered more, and to establish whether geographical disparities persist. Notably, the countries that are most actively developing CAR T-cell trials are those with less emphasis on quality of life, which is an alarming finding.

This study revealed a lack of consensus and heterogeneity in HRQoL assessment among patients treated with CAR T cells for hematologic malignancies. It is therefore crucial to establish a consensus on collecting HRQoL data in clinical trials and in real life, enabling all patients worldwide to benefit from these therapeutic innovations.

All authors have made substantial contributions to all of the following: (1) the conception and design of the study, acquisition of data, analysis and interpretation of data, (2) drafting the article or revising it critically for important intellectual content, (3) final approval of the version to be submitted.