LBCL is the most common subtype of non-Hodgkin lymphoma worldwide, accounting for roughly 30 to 60 percent of cases, and diffuse large B cell lymphoma represents about one-third of the estimated 79,320 non-Hodgkin lymphoma cases projected in the US in 2026. Approximately half to 60 percent of patients achieve a durable response to first-line immunochemotherapy with rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone, but the remainder become refractory or relapse early. Historically, transplant-eligible patients failing first-line therapy received salvage immunochemotherapy followed by high-dose chemotherapy and autologous hematopoietic stem cell transplantation. Yet about half of patients intended for that transplant cannot complete it because their disease responds insufficiently to salvage treatment, and those ineligible face a dismal median overall survival of just four months.

Two randomized trials tested CAR T cell therapy against standard of care in this population. The TRANSFORM study showed that liso-cel, and the ZUMA-7 study showed that axi-cel, each outperformed standard of care in patients with primary refractory or early relapsed LBCL within 12 months of first-line chemoimmunotherapy, with durable efficacy during follow-up. Because the two products were never compared directly in a single randomized trial, the authors of the new analysis, led by Kristen A. Deger and colleagues, used an indirect treatment comparison to estimate how they stack up against each other in both clinical benefit and economic value.

The team built a partitioned survival model in which patients occupy one of three health states over time: event-free, post-event, or death. Because some patients receiving second-line CAR T cell therapy appear to achieve long-term remission, with survival curves that plateau, the researchers employed mixture cure modeling to extrapolate event-free survival and overall survival beyond the observed trial follow-up. This statistical framework assumes the population consists of a cured fraction, whose mortality resembles that of the age- and sex-matched general US population, and a noncured fraction, whose survival is predicted by parametric distributions chosen for both statistical fit and clinical plausibility. Individual patient-level data from TRANSFORM, with a median follow-up of 33.9 months, and reconstructed data from ZUMA-7, with a median follow-up of 47.2 months, anchored the projections. The comparative effect of axi-cel versus liso-cel was derived using the Bucher method, which leverages the standard-of-care arm common to both trials.

The model was deliberately granular on the economic side. Costs included CAR T cell pretreatment, comprising leukapheresis, bridging therapy, and lymphodepleting chemotherapy; drug acquisition and administration in inpatient or outpatient settings; hospitalization after infusion; adverse event management; monitoring; subsequent third-line treatments; and end-of-life care, all inflated to 2026 US dollars. Notably, the analysis applied microcosting to the toxicities of special interest in CAR T cell therapy: cytokine release syndrome, neurological events, and hypogammaglobulinemia, capturing intensive care use separately from the routine post-infusion hospital stay to avoid double counting. A key practical difference between the products emerged here: in TRANSFORM, 21.3 percent of patients received their infusion as outpatients and 78.7 percent as inpatients with a 14-day stay, whereas in ZUMA-7 only 3.0 percent were infused as outpatients and the reported inpatient stay averaged 15 days.

Safety differences also fed into the quality-of-life estimates. The model incorporated adverse events of grade 3 or higher occurring in at least 5 percent of patients per the US prescribing information, plus all-grade cytokine release syndrome, neurological events, and hypogammaglobulinemia regardless of incidence. Health state utilities were derived from EQ-5D-5L data collected in TRANSFORM and analyzed with a mixed-effects model for repeated measures, supplemented by a published time trade-off study for disutilities that the trial data were too small to estimate, such as those for cytokine release syndrome and neurological events. Liso-cel’s more favorable safety profile, with lower rates of these toxicities, contributed a modest 0.05 QALY advantage of its own.

The base-case results showed a narrow but consistent edge for liso-cel. Projected five-year overall survival was 59 percent with liso-cel versus 58 percent with axi-cel, and over a lifetime horizon liso-cel yielded 10.78 life-years and 9.49 QALYs compared with 10.51 and 9.18, respectively. Total costs were $826,598 for liso-cel versus $835,996 for axi-cel. Although liso-cel carried a higher drug acquisition cost, this was more than offset by savings of $15,074 in CAR T cell administration and hospitalization, $11,109 in adverse event management, and $2,864 in subsequent treatment. The resulting incremental net monetary benefit was a positive $40,210 at the widely used willingness-to-pay threshold of $100,000 per QALY gained.

Sensitivity analyses revealed both the robustness and the fragility of the conclusion. Deterministic sensitivity analysis identified the overall survival odds of cure for liso-cel as the single dominant driver of the results. In the probabilistic sensitivity analysis of 3,000 simulations, results clustered predominantly in the southeast and southwest quadrants of the cost-effectiveness plane, indicating greater confidence that liso-cel is less costly than axi-cel than that it is more effective. Liso-cel was more effective in 58 percent of simulations and less costly in 42 percent, and its probability of being cost-effective at the $100,000 per QALY threshold was 62 percent. Scenario analyses showed that assumptions about third-line treatment mattered most: assuming all patients progressing received the relatively expensive polatuzumab plus bendamustine and rituximab regimen cut incremental costs by 40 percent in liso-cel’s favor, since modeled patients on liso-cel reached third-line treatment more slowly.

The authors acknowledge important limitations. Long-term overall survival after second-line CAR T cell therapy remains uncertain, and only three distributions were deemed both statistically and clinically plausible for extrapolating the survival of the noncured population, although the projections were consistent with Kaplan-Meier estimates from TRANSFORM and bounded below by longer-term standard-of-care data from the CORAL trial. The TRANSFORM utility analysis was limited by a small sample size and low compliance, and residual differences between the trials, including bridging therapy rules and infusion settings, could bias the indirect comparison. Trial data may also not reflect real-world outcomes, and the authors note that demographic differences in access and outcomes warrant stratified investigation in future studies.

Within those boundaries, the analysis extends a previous finding that liso-cel was cost-effective versus axi-cel in the third-line or later setting, and in both cases the cure fraction drove the economics. The conclusion, as framed by the study, is that from a US commercial third-party payer perspective, patients treated with liso-cel as second-line therapy for transplant-eligible LBCL are estimated to gain slightly more life-years and QALYs at a lower total cost than those treated with axi-cel, an estimate that will require confirmation against real-world data as they accumulate.

Subject of Research: Cost-Effectiveness of Lisocabtagene Maraleucel Versus Axicabtagene Ciloleucel as Second-Line Treatment for Large B Cell Lymphoma in the United States

Article Title: Cost-Effectiveness of Lisocabtagene Maraleucel Versus Axicabtagene Ciloleucel as Second-Line Treatment for Large B Cell Lymphoma in the United States

Article References: Deger, K. A., Liu, F. F., Klijn, S. L., Franco-Villalobos, C., Sorensen, S. V., & Lunning, M. (2026). Cost-Effectiveness of Lisocabtagene Maraleucel Versus Axicabtagene Ciloleucel as Second-Line Treatment for Large B Cell Lymphoma in the United States. Advances in Therapy. https://doi.org/10.1007/s12325-026-03827-6

Image Credits: AI Generated

DOI: 10.1007/s12325-026-03827-6

Keywords: Cost-Effectiveness, Lisocabtagene, Maraleucel, Versus, Axicabtagene, Ciloleucel, Second-Line, Treatment, Large, Cell, Lymphoma, United