Key Ideas

CAR-T cell therapy is the established standard second-line treatment for primary refractory or early relapsed diffuse large B-cell lymphoma, supported by mature, practice-defining trial data. Bispecific antibody-chemotherapy combinations are now emerging as a competitive alternative, particularly for patients who cannot navigate the three-to-five-week manufacturing wait that CAR-T requires or who lack access to a cellular therapy center.

A Representative Case

A 63-year-old woman with no significant comorbidities, good performance status, and strong social support is diagnosed with germinal-center-B-cell-subtype diffuse large B-cell lymphoma. She is treated with R-CHOP; an interim PET scan after three cycles shows response, but the end-of-treatment PET scan after six cycles reveals new areas of lymphadenopathy, and biopsy confirms persistent disease. This is a textbook presentation of primary refractory diffuse large B-cell lymphoma, and it sets up one of the more consequential decisions in modern lymphoma care: CAR-T, or a bispecific-based alternative.

The Trials That Made CAR-T the Standard

Two pivotal randomized trials established cellular therapy as the preferred second-line option for primary refractory or early relapsed (within 12 months) large B-cell lymphoma: ZUMA-7, testing axicabtagene ciloleucel, and TRANSFORM, testing lisocabtagene maraleucel. Both randomized patients to CAR-T versus standard-of-care salvage chemotherapy followed by autologous stem cell transplant in responders, and both were positive.

Complete response rates were 65% with axicabtagene ciloleucel and 74% with lisocabtagene maraleucel. Median event-free survival was 10.8 months with axicabtagene ciloleucel and 29.5 months with lisocabtagene maraleucel (each compared with only a few months in the respective standard-of-care arms). Four-year overall survival with axicabtagene ciloleucel reached 54.6%, and three-year overall survival with lisocabtagene maraleucel reached 63%. These are not directly comparable numbers, since the trials differ in design, follow-up duration, and crossover rules, but taken together they establish that roughly half of this historically difficult-to-treat population is now being durably controlled, and in many cases likely cured, by second-line cellular therapy. Toxicity differed meaningfully between products: cytokine release syndrome of any grade occurred in 92% of axicabtagene ciloleucel patients versus 49% with lisocabtagene maraleucel, with grade 3 or higher rates of 6% and 1% respectively, a distinction that matters when selecting between the two approved products for an individual patient.

The Real Gaps in the CAR-T Pathway

CAR-T's efficacy does not eliminate its practical limitations. From apheresis to product infusion typically takes three to five weeks for manufacturing, a window during which a meaningful share of patients, some estimates suggest around 20%, never make it to infusion because their disease cannot wait or progresses too quickly during that interval. Salvage regimens used to control disease during this bridging period can themselves affect T-cell fitness, complicating the product being manufactured. Using bendamustine or other CD19-targeted therapies prior to apheresis specifically raises concern, since CAR-T products target CD19 directly; the general approach is to complete apheresis first when possible, then use whatever bridging strategy is needed afterward with less concern about compromising the collected T cells. Whether bispecific antibody exposure before apheresis meaningfully burns out T cells is an open question in the lymphoma space specifically, distinct from the more established long-term bispecific exposure data available in multiple myeloma, where the treatment paradigm differs considerably. Beyond manufacturing timelines, access itself remains a persistent barrier: reaching CAR-T generally requires proximity to, or the social support to travel to, a specialized academic center, which a meaningful share of eligible patients simply do not have.

STARGLO and SUNMO: The Emerging Bispecific Alternative

Two trials have defined the case for bispecific-chemotherapy combinations in this same primary refractory and early relapse population. STARGLO combined glofitamab, a CD3xCD20 bispecific antibody, with rituximab as a comparator, in relapsed or refractory diffuse large B-cell lymphoma patients ineligible for transplant, delivered as fixed-duration treatment over roughly one year. The combination produced a 68% overall response rate and 58% complete response rate, with efficacy holding up specifically in the primary refractory subgroup, and progression-free and overall survival curves that, while not yet mature, are beginning to show a plateau out to several years of follow-up, a pattern reminiscent of what was seen with ZUMA-7 and TRANSFORM.

SUNMO tested a different bispecific antibody, mosunetuzumab, in combination with an antibody-drug conjugate against a chemoimmunotherapy comparator, again in transplant-ineligible patients, including some with transformed and double-hit lymphoma histologies not always captured in other trials. Overall response rate reached 70%, with a 51% complete response rate, and a meaningful share of enrolled patients had primary refractory disease specifically, with encouraging signal in that subgroup as well. Progression-free survival curves here also show an early plateau, around 40% out to three years of follow-up, though this data, like STARGLO's, still needs longer maturation. Both regimens showed cytokine release syndrome and neurologic toxicity that was considerably easier to manage than what is typically seen with CAR-T products.

These results are compelling enough that bispecific-chemotherapy combinations should be considered an option in select clinical situations, even though NCCN has not yet endorsed them over CAR-T as the current second-line standard for this population, reflecting how much less mature this data still is compared with the CAR-T trials.

Choosing Between the Two Approaches

Several practical factors favor one strategy over the other in a given patient. Fitness or performance status rarely disqualifies someone from CAR-T outright, since eligibility criteria are broad, but for the borderline patient, a bispecific-based regimen may be the more appropriate choice. Disease urgency matters considerably: patients who cannot tolerate the three-to-five-week apheresis-to-infusion window are often better served by a chemotherapy-bispecific combination that can start immediately. Toxicity risk tolerance is another consideration, since CAR-T carries meaningfully higher rates of cytokine release syndrome and neurologic toxicity than the bispecific combinations described here. Access to a CAR-T center is, practically speaking, one of the largest drivers of this decision in real-world practice. Patient preference for a one-time therapy also matters: CAR-T offers a one-and-done treatment course, while bispecific combinations in this setting currently run up to a year of continuous treatment. Finally, for patients and clinicians who prioritize the most mature, longest-follow-up data available, that consideration still favors CAR-T today, though the bispecific data is expected to continue maturing favorably.

For the case described above, a fit 63-year-old with strong social support living near a cellular therapy center, these features favor proceeding to CAR-T. A different patient profile, one further from a treatment center, less fit, or facing rapidly progressing disease that cannot wait for manufacturing, would more reasonably point toward a bispecific-chemotherapy combination instead.

For Patients

When diffuse large B-cell lymphoma does not fully respond to initial chemotherapy, often called primary refractory disease, two main treatment paths have emerged as strong options. CAR-T cell therapy, which reprograms a patient's own immune cells in a lab before infusing them back into the body, has the longest track record and has cured roughly half of patients in this difficult situation in large clinical trials, an outcome that would not have been possible with older treatment approaches.

CAR-T does require several weeks between the initial cell collection and the actual treatment, time needed to manufacture the product, and not every patient's disease can wait that long, or every patient can access a specialized center offering this therapy. For those situations, newer combinations of antibody-based treatments called bispecific antibodies with other cancer drugs are showing promising results, can typically start right away, and generally cause fewer of the serious side effects associated with CAR-T, such as high fevers and confusion in the days after treatment.

Patients with lymphoma that has not responded to initial treatment should ask their oncologist directly about both CAR-T and newer bispecific antibody options, since the right choice depends heavily on individual factors: how quickly the disease is progressing, overall health, proximity to a specialized treatment center, and personal preference for a one-time treatment versus an extended course. Both approaches represent progress in a disease area where options were once far more limited.

Key Takeaways

  • ZUMA-7 (axicabtagene ciloleucel) and TRANSFORM (lisocabtagene maraleucel) established CAR-T as the standard second-line therapy for primary refractory or early relapsed diffuse large B-cell lymphoma, with four-year overall survival of 54.6% and three-year overall survival of 63% respectively.

  • Roughly 20% of patients referred for CAR-T never reach infusion due to the three-to-five-week manufacturing window, disease progression during that time, or access barriers, underscoring limitations in an otherwise highly effective therapy.

  • Bispecific antibody-chemotherapy combinations, particularly glofitamab-based (STARGLO) and mosunetuzumab-based (SUNMO) regimens, are showing strong response rates and early survival signals in the same primary refractory population, with substantially lower rates of cytokine release syndrome and neurologic toxicity than CAR-T.

  • NCCN has not yet endorsed bispecific combinations over CAR-T as the second-line standard, reflecting the relative immaturity of that data, but these combinations represent an option for patients who cannot access or wait for CAR-T.

  • Choosing between CAR-T and a bispecific-based approach depends on disease urgency, fitness, toxicity tolerance, access to a cellular therapy center, and patient preference for a one-time versus extended treatment course.

References

  1. Locke FL, Ahmed S, Neelapu SS, et al. Axicabtagene ciloleucel as second-line therapy for large B-cell lymphoma (ZUMA-7). New England Journal of Medicine. 2022.

  2. Westin JR, Oluwole OO, Kersten MJ, et al. Survival with axicabtagene ciloleucel in large B-cell lymphoma. New England Journal of Medicine. 2023.

  3. Abramson JS, Solomon SR, Arnason J, et al. Lisocabtagene maraleucel as second-line therapy for large B-cell lymphoma: primary analysis of the phase 3 TRANSFORM study. Blood. 2023.

  4. Kamdar M, Solomon SR, Arnason J, et al. Lisocabtagene maraleucel vs standard of care as second-line treatment for large B-cell lymphoma: 3-year follow-up from the randomized, phase 3 TRANSFORM study. Journal of Clinical Oncology. 2024.

  5. Abramson JS, Ku M, Hertzberg M, et al. Glofitamab plus gemcitabine and oxaliplatin (Glofit-GemOx) versus rituximab-GemOx for relapsed or refractory diffuse large B-cell lymphoma (STARGLO): a global phase 3, randomised, open-label trial. Lancet. 2024.

  6. Budde LE, Zhang H, Kim WS, et al. Mosunetuzumab plus polatuzumab vedotin in transplant-ineligible refractory/relapsed large B-cell lymphoma: primary results of the phase 3 SUNMO trial. Journal of Clinical Oncology. 2025.