Clinicians are under-recording the substantial early side-effect burden caused by CAR-T-cell therapy, new research suggests.

The Italian study, which was part-funded by Novartis Farma Italy, showed that 12 out of 19 assessed symptoms were reported by at least 50% of CAR-T recipients ten days after treatment. 

However, these side effects were frequently not recorded by clinicians, with all any-grade events, except fatigue, missed at least 60% of the time.

Moderate to severe symptoms were also frequently under-reported, with clinicians failing to record all patient-reported cases of difficult swallowing and general pain, as well as more than 90% of dizziness, muscle pain and nausea.

“Our findings show that the immediate post-CAR T-cell therapy period is characterised by a substantial patient-experienced symptom burden that is not fully captured by clinician-only toxic effect reporting,” wrote the authors, led by Dr Fabio Efficace of the Italian Group for Adult Haematologic Diseases (GIMEMA) in Rome, Italy.

“Routine integration of patient-reported outcome measures (eg, PRO-CTCAE) is needed to improve timely recognition of supportive care needs, strengthen patient counselling on expected symptom trajectories, and enable more patient-centred monitoring strategies,” they suggested.

The study, published in The Lancet Haematology [link here], involved 170 patients with aggressive B-cell lymphoma (median age 61 years, 69% male) who underwent CAR T-cell therapy with either axicabtagene ciloleucel (axi-cel, 58%), tisagenlecleucel (tisa-cel, 23%), or brexucabtagene autoleucel (brexu-cel, 19%).

Most patients were diagnosed with diffuse large B-cell lymphoma (69%), followed by mantle cell lymphoma (19%) and primary mediastinal large B-cell lymphoma (12%).

Ten days after treatment, 12 of the 19 symptoms on the Common Terminology Criteria for Adverse Events (CTCAE) questionnaire were reported by at least half of respondents, specifically:

• fatigue (87% of patients)
• decreased appetite (83%)
• insomnia (79%)
• chills (73%)
• diarrhoea (70%)
• sadness (57%)
• general pain (55%)
• dizziness (55%)
• joint pain (52%)
• muscle pain (51%)
• headache (51%)
• impaired concentration (51%). 

At the same time point, at least 30% of patients reported moderate to severe fatigue (55%), decreased appetite (53%), diarrhoea (45%), chills (35%), and insomnia (31%).

Females were significantly more likely than males to experience impaired concentration or hair loss of any grade, as well as moderate to severe fatigue, general pain, dizziness, joint pain, muscle pain and memory difficulties.

However, age did not impact symptom burden, nor did the type of CAR-T product.

Clinicians frequently failed to record symptoms that were reported by patients, with only fair to moderate concordance overall.

Among any grade symptoms, hair loss was the most frequently under-reported symptom with 91% of patients (19/21) reporting the symptom without their physician recording it, while fatigue was the least under-reported at 43%.

Under-reporting was also high for swelling (83%), musculoskeletal symptoms, such as muscle pain and joint pain (both 73%), and gastrointestinal symptoms including difficulty swallowing (78%), decreased appetite (77%), nausea (70%), and diarrhoea (71%). 

Clinicians frequently failed to recognise neuropsychiatric symptoms reported by patients, most notably dizziness (89%), discouragement (82%) and insomnia (71%).

Moderate to severe symptoms were also frequently under-reported, with 100% of patient-reported cases of difficulty swallowing (11/11) and general pain (17/17) missed by clinicians and under-reporting of dizziness (94%), muscle pain (94%) and nausea (92%) also high.

The data also showed that male sex (p=0.048) and higher baseline EORTC QLQ-C30 physical functioning score (p=0.0091) were associated with lower symptom burden.

“Clinician-assessed variables (performance status), disease status, or other biomedical variables reflecting clinical history were not associated with day 10 symptom burden,” the authors noted.

“Our data suggest that patient-reported fitness might be more sensitive than traditional clinical indicators in capturing susceptibility to early toxic effects,” they added.

“Incorporating pre-infusion patient-reported physical functioning could help identify the most susceptible patients at increased risk of early symptom burden, thereby supporting implementation of tailored supportive care strategies,” they concluded.