CD19 CAR T-cell therapy for relapsed/refractory diffuse large B-cell lymphoma in a nonagenarian patient

Abstract

Chimeric antigen receptor (CAR) T-cell therapy has revolutionized the management of relapsed/refractory large B-cell lymphoma. However, evidence in the ultra-elderly population remains scarce. We report a 90-year-old woman with relapsed/refractory diffuse large B-cell lymphoma transformed from follicular lymphoma who received lisocabtagene maraleucel. Despite advanced age, preserved cognition, intact instrumental activities of daily living, and acceptable organ reserve supported her candidacy for CAR T-cell therapy. Cytokine release syndrome (CRS) developed as grade 1 (day 1, 38.0 °C) and grade 2 (day 3, hypoxemia to 6 L/min), which resolved after early tocilizumab and dexamethasone. No immune effector cell–associated neurotoxicity syndrome or infectious complications occurred. Grade 2–4 cytopenia persisted transiently but was managed with supportive care, and she was discharged on day 28. To our knowledge, this represents the first reported case of a nonagenarian Asian patient successfully treated with CAR T-cell therapy, highlighting the feasibility and tolerability of this approach and emphasizing the importance of geriatric assessment and prompt CRS management for safe delivery in the very elderly.

INTRODUCTION

Diffuse large B-cell lymphoma (DLBCL) predominantly affects older adults, with a median age at diagnosis over 65 years. Chimeric antigen receptor (CAR) T-cell therapy has transformed the management of relapsed/refractory DLBCL, but pivotal trials and most real-world reports focus on younger, medically fit patients.^1,2 Emerging evidence suggests that selected elderly and very elderly patients, including octogenarians, can achieve comparable efficacy with acceptable safety.^3–5 Geriatric assessment helps identify vulnerabilities that guide individualized treatment and predict safety in older lymphoma patients.^6,7 However, in the context of CAR T-cell therapy, few studies have systematically evaluated safety using geriatric assessment, so CAR T-cell therapy is rarely performed in the very elderly. We describe a 90-year-old patient with relapsed/refractory DLBCL treated with anti-CD19 CAR T-cell therapy, underscoring the feasibility and the importance of individualized, geriatric assessment–guided decision-making beyond chronological age.

CASE REPORT

An 81-year-old woman was diagnosed in August 2016 with follicular lymphoma (FL) grade 3A, stage IV (FLIPI 1: low risk; FLIPI-2: intermediate risk). Immunohistochemistry (left axillary node) showed CD20+, CD10+, BCL6+, Ki-67 ~60%; negative for BCL2, CD5, cyclin D1, and MUM1. Baseline positron emission tomography–computed tomography (PET-CT) demonstrated a 17-mm left axillary node (a maximum standardized uptake value (SUVmax) 10.61). Given low tumor burden without B symptoms or organ dysfunction, watchful waiting was chosen. In February, as lower back pain worsened, PET-CT (Figure 1a) showed soft-tissue masses in the left axilla, para-aortic region, and sacrum with high uptake (SUVmax 12.68) and multiple osseous involvements. Bendamustine-rituximab ×6 achieved complete metabolic remission (CMR); however, in August 2019 PET-CT revealed new gastric wall uptake (SUVmax 19.0). Biopsy confirmed transformation to DLBCL (non–germinal center B-cell subtype) with CD20+ and BCL2+, Ki-67 about 50%, and negative CD10, CD5, BCL6, and MUM1. She received R-CHOP (rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone) ×6 followed by rituximab ×2, achieving CMR. In May 2024, left leg edema prompted PET-CT (Figure 1b) showing lymphadenopathy from the right posterior diaphragmatic crus through the para-aortic to iliac regions and at L5 (SUVmax 13.56). Treating as recurrent FL, rituximab-lenalidomide was started, but the disease progressed. Although no repeat biopsy was performed due to the absence of easily accessible lesions, the rapid clinical progression suggested relapse as DLBCL rather than FL; therefore, epcoritamab was administered for four cycles. In October 2024, PET-CT (Figure 1c) demonstrated enlargement of para-aortic/iliac disease (SUVmax up to 13.58), new left supraclavicular nodes, and skull uptake (SUVmax 4.58), consistent with progressive metabolic disease (PMD).

Fig. 1.

PET-CT evaluation before and after CAR T-cell infusion

(a) February 2019: Baseline scan before initial therapy, showing soft tissue masses in the left axilla, para-aortic region, and sacrum with high uptake (SUVmax 12.68) and osseous involvement.

(b) May 2024: Lymphadenopathy in the diaphragmatic crus, para-aortic to iliac regions, and L5 vertebra (SUVmax 13.56).

(c) October 2024: Before CAR T-cell bridging therapy, demonstrating enlarged para-aortic and iliac lesions (SUVmax 13.58), new left supraclavicular lymph nodes, and skull uptake (SUVmax 4.58), consistent with PMD.

(d) April 2025: Residual uptake in pelvic and left supraclavicular lesions after CAR T-cell therapy, corresponding to PMR.

(e) July 2025: Re-enlargement of pelvic and left supraclavicular lesions (SUVmax 14.26), consistent with PMD.

At CAR T evaluation, she was 141 cm/38 kg (body mass index 19.1 kg/m²), Eastern Cooperative Oncology Group performance status 2, cognitively intact, Hematopoietic Cell Transplantation–Specific Comorbidity Index (HCT-CI) 3 (remote brain-tumor surgery in remission), and no cardiac or pulmonary dysfunction with creatinine clearance 16 mL/min. Instrumental activities of daily living (IADL) score was 8, her Charlson Comorbidity Index was 2 (intermediate risk),⁸ Geriatric Nutritional Risk Index indicated moderate risk,⁹ and the G8 screening tool score was 8, classifying her as frail.¹⁰ In the comprehensive geriatric assessment, no clinically significant depressive symptoms nor cognitive impairment, or deficits in social support were identified, and she remained independent in basic activities of daily living, which together supported the decision to proceed with CAR T-cell therapy despite the frail G8 score. Bridging rituximab with dexamethasone, etoposide, ifosfamide, and carboplatin (R-DEVIC; 50% dose, adjusted for age and renal impairment) was given in November 2024; leukapheresis for lisocabtagene maraleucel (liso-cel) followed in December, 48 days after the last epcoritamab. At apheresis, laboratory values were as follows: white blood cell count 3,300/µL, lymphocyte count 842/µL, hemoglobin 9.5 g/dL, platelet count 200 × 10³/µL, LDH 238 U/L, and C-reactive protein (CRP) 0.04 mg/dL. A second R-DEVIC yielded a partial metabolic response in abdominal nodes. Before lymphodepleting chemotherapy, laboratory values were as follows: white blood cell count 3,200/µL, lymphocyte count 304/µL, CD4-positive T-cell count 89/µL, hemoglobin 9.1 g/dL, platelet count 223 × 10³/µL, LDH 269 U/L, ferritin 95 ng/mL, and CRP 0.51 mg/dL. In January 2025, fludarabine 18.75 mg/m²/day (days -4 to -2) and cyclophosphamide 300 mg/m² (days -4 to -3) were administered; liso-cel 1.35 × 10⁶ cells/kg was infused on day 0 when she was 90 years old (Table 1). The Immune Effector Cell–Associated Encephalopathy (ICE) score was 10. The CAR-HEMATOTOX score before lymphodepleting chemotherapy was 0. Grade 1 cytokine release syndrome (CRS) (day 1, 38.0 °C) responded to tocilizumab; grade 2 CRS (day 3, hypoxemia to 6 L/min) resolved with tocilizumab plus dexamethasone (10 mg q12h), with grading defined according to the American Society for Transplantation and Cellular Therapy consensus criteria. Oxygen and steroids were discontinued by day 6, and no immune effector cell–associated neurotoxicity syndrome (ICANS) was observed; the ICE score remained 10 after infusion. Grade 2–4 neutropenia persisted through day 27, requiring intermittent granulocyte colony-stimulating factor; she remained red-cell transfusion–dependent through day 39 and platelet-dependent through day 44. She had no infectious complications and was discharged on day 28. Serum IgG was 446 mg/dL pre-infusion, 302 mg/dL at 1 month, and 497 mg/dL at 2 months. A follow-up PET-CT in April 2025 revealed increased uptake limited to a single intra-abdominal lesion, with markedly reduced overall disease burden compared to baseline prior to CAR T-cell therapy (Figure 1d). Because no new sites were detected and most sites showed ongoing regression, this finding was interpreted as indolent progression, and the patient was managed with close observation. On observation, back pain, LDH, and sIL-2R rose; July 2025 PET-CT confirmed re-enlargement with high uptake (SUVmax 14.26), consistent with PMD (Figure 1e). Salvage polatuzumab vedotin-rituximab-cyclophosphamide-prednisone (Pola-R-CP; 50% cyclophosphamide/prednisone) was initiated.

Table 1. Early clinical course of cytokine release syndrome after lisocabtagene maraleucel infusion (graded per ASTCT 2019 criteria).

Day (from infusion)	Temp (°C)	Blood pressure	O2 requirement	ASTCT CRS grade	Key interventions & events	Notes
0	–	Stable	None	0	Liso-cel infused (1.35 × 106 cells/kg)	Age 90 at infusion
1	38.0	Stable	None	1	Tocilizumab 8 mg/kg administered	Fever onset
2	–	Stable	None	0–1	Observation; supportive care	No new interventions documented
3	37.6	Stable	Up to 6 L/min	2	Tocilizumab 8 mg/kg; dexamethasone 10 mg q12h initiated	Hypoxemia without vasopressors
4–5	–	Stable	Weaning	1	Continue steroids; taper oxygen as tolerated	–
6	–	Stable	Off oxygen	0–1	Steroids discontinued	Clinical improvement
28	–	–	–	–	Discharged	No ICANS or infections

Abbreviations: ASTCT, American Society for Transplantation and Cellular Therapy; CRS, Cytokine release syndrome; ICANS, Immune effector cell–associated neurotoxicity syndrome; O₂, Oxygen; TCZ, Tocilizumab; DEX, Dexamethasone; q12h, Every 12 hours; liso-cel, Lisocabtagene maraleucel

DISCUSSION

This case illustrates the feasibility and tolerability of liso-cel in a nonagenarian with multiply relapsed diffuse large B-cell lymphoma after transformation from follicular lymphoma. To our knowledge, this is the oldest reported case of an Asian patient successfully treated with CAR T-cell therapy. Despite her advanced age, preserved cognitive function, intact IADL, and acceptable organ reserve supported treatment selection. CRS was manageable, and she was discharged on day 28 without major complications.

Recent multicenter real-world studies provide convergent evidence that anti-CD19 CAR T-cell therapy can benefit older adults, including advanced age groups (Table 2).^3,5,11–13 Across four cohorts, median ages ranged from 71 to 82 years; axi-cel, tisa-cel, and liso-cel were commonly used. Complete remission rates remained robust (45–71%), with 12-month progression-free and overall survival rates of 32–48% and 52–70%, respectively. Rates of grade ≥3 CRS and ICANS were generally low (2–15% and 2.5%–26%, respectively), fatal toxicity was infrequent, and 12-month non-relapse mortality was typically ≤10%. Importantly, relapse rates at 12 months were not clearly higher than in younger comparators. Collectively, these data suggest that chronological age alone should not preclude CAR T-cell therapy when fitness and comorbidities are appropriately evaluated and managed. For older adults, recent evidence demonstrates that comprehensive geriatric assessment—not just age—improves risk stratification and decision-making for CAR T-cell therapy. Multidisciplinary evaluation of physical function, nutrition, cognition, and comorbidities allows identification of patients most likely to benefit, while those with significant vulnerabilities have higher toxicity and worse outcomes even after optimization.¹⁴ Additionally, comorbidity indices such as the Cellular Therapy Comorbidity Index are independently associated with survival after CAR-T, supporting the integration of both geriatric assessment and comorbidity scoring into clinical practice to guide therapy selection.¹⁵ In addition, in this case, liso-cel was selected because of its relatively favorable safety profile, particularly the lower incidence of severe cytokine release syndrome and neurotoxicity compared with some other CD19-directed CAR T-cell products.^16,17 In very elderly patients, reducing the risk of high-grade CRS and ICANS is particularly important, and liso-cel is therefore an attractive option when available.

Table 2. Clinical outcomes of anti-CD19 CAR T-cell therapy in elderly patients with B-cell lymphoma.

Reference	[5]	[11]	[3]	[12]	[13]
Indication	LBCL	B-cell lymphoma	DLBCL	LBCL	LBCL
Median age (range), years	74 (53–84)	82 (80–89)	76.2 (70–)	73 (70–81)	71 (65–89)
Male sex (%)	61%	73.90%	39%	55%	63%
Median number of lines of prior systemic therapy (range)	1	2 (1–7)	2 (2–>5)	2 (2–6)	2 (1–8)
Median follow up, months	13.0	9.2	7	12.2	18.3
Products	-
- Axi-cel	100%	46.6%	19.5%	59.1%	58.4%
- Liso-cel	-	28.4%	-	-	16.4%
- Tisa-cel	-	14.8%	80.5%	40.8%	25.2%
- Brexu-cel		10.2%	-	-	-
CR rate (%)	54%	71.4%	46%	45%	62%
PFS (12 months)	45.8%	47.6% †	32%	35.9%	44%
OS (12 months)	70.0%	61.2% †	69%	52.1%	61%
NRM (3 months)	-	1.4% †	0%	3%	8.9%
NRM (12 months)	-	11.6% †	-	6%	10.9% (day 180)
Relapse (12 months)	-	40.8% †	-	-	35.5% (day 180)
Any grade CRS (%)	38%	77.3%	68.3%	84.5%	75%
Grade ≥3 CRS (%)	2%	5.7%	9.8%	15%	7%
Any grade ICANS (%)	31%	58.0%	27.5%	42%	52%
Grade ≥3 ICANS (%)	5%	18.1%	2.5%	17%	26%

†indicates data specific to patients with DLBCL and transformed FL.

Abbreviations: DLBCL, diffuse large B-cell lymphoma; LBCL, large B-cell lymphoma; Axi-cel, axicabtagene ciloleucel; Liso-cel, lisocabtagene maraleucel; Tisa-cel, tisagenlecleucel; Brexu-cel, brexucabtagene autoleucel; CR, complete response (=complete remission); PFS, progression-free survival; OS, overall survival; NRM, non-relapse mortality; CRS, cytokine release syndrome; ICANS, immune effector cell–associated neurotoxicity syndrome

Our patient’s clinical course also underscores the value of early CRS management. Emerging cohort and prospective data indicate that prompt administration of tocilizumab, with early corticosteroids when warranted, reduces progression to high-grade CRS, intensive care utilization, and treatment-related mortality without compromising antitumor efficacy, neurotoxicity rates, or infection risk.^18,19 In this case, early tocilizumab for initial fever and timely addition of dexamethasone at grade 2 CRS achieved rapid control, with no ICANS or infectious complications and timely discharge. However, this case represented an early relapse after CAR T-cell therapy. Recent studies have reported that CAR T-cell therapy following prior bispecific antibody (BsAb) exposure can achieve comparable response rates to those observed in BsAb-naïve patients.²⁰ However, when administered shortly after BsAb treatment, there may be an increased risk of treatment resistance due to T-cell exhaustion.²¹ In addition, the therapeutic efficacy of CD19-directed CAR T-cell therapy depends on the expression of the CD19 antigen on tumor cells. Loss or downregulation of CD19 expression—so-called antigen escape—represents a major mechanism of treatment failure and disease relapse.²² In retrospect, given the absence of an easily accessible lesion for biopsy, CAR T-cell therapy was administered without histologic confirmation, which represents a limitation of this case.

In conclusion, a comprehensive geriatric assessment—not age alone—should guide CAR T-cell candidacy in very elderly patients with relapsed/refractory large B-cell lymphoma. When physiologic reserve is adequate and CRS is recognized and treated promptly with evidence-based algorithms, CAR T-cell therapy can be delivered safely even in nonagenarians. Prospective registries and trials incorporating standardized geriatric and comorbidity metrics (e.g., IADL, HCT-CI) are warranted to refine patient selection and supportive-care strategies in this growing population.

DATA AVAILABILITY STATEMENT

All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.