Key Ideas

Second-line care for relapsed large B-cell lymphoma has moved decisively away from salvage chemotherapy and transplant toward CAR T-cell therapy for patients with early relapse. For patients who cannot pursue curative-intent treatment, bispecific antibodies and antibody-drug conjugates now offer durable, chemotherapy-free options. Transplant ineligibility no longer means CAR T ineligibility, and the growing menu of effective agents has shifted the central challenge from lack of options to choosing among them.

From More Chemotherapy to CAR T: How Second-Line Care Has Changed

The treatment landscape for relapsed or refractory diffuse large B-cell lymphoma (DLBCL) has changed substantially in a short period. For patients with early relapse, salvage chemotherapy and transplant are no longer the default; CAR T-cell therapy has taken that role. For patients who are not eligible for curative-intent treatment with CAR-T, effective options, most of them bispecific antibody combinations, are now available. Among patients eligible for both bispecific therapy and CAR-T, CAR-T therapy remains the priority when feasible due to its potential to cure, and effective off-the-shelf options remain available for those who relapse afterward.

Looking ahead, novel agents are moving earlier in the treatment course. Polatuzumab vedotin is already used in the first line, and bispecific antibodies and CAR T-cell therapy are increasingly being tested there as well. That raises new questions: what will remain effective at relapse once these agents are used earlier, whether molecular measures such as measurable residual disease or circulating tumor DNA might identify failure before radiographic relapse, whether DLBCL should be approached as multiple molecular subtypes rather than one disease, and how newer, more sophisticated immune platforms, including CD19 and CD22 bispecific antibodies, trispecific antibodies, dual-targeted constructs, and allogeneic cell therapies, will fit into practice.

Why Early Relapse Still Anchors the Algorithm

First-line chemoimmunotherapy remains curative for most patients with DLBCL, but relapse remains a real problem, and most relapses occur within the first two years, a pattern consistent with Swedish lymphoma registry data and the GOYA study. That distinction matters clinically. A few years ago, patients who relapsed after first-line chemoimmunotherapy were generally offered more chemotherapy in the hope of a chemosensitive response, followed by transplant with curative intent; outcomes for chemo-resistant patients grew progressively worse under that model.

The first question in a relapsed patient is timing: did relapse occur within 12 months of initial treatment, or later? If relapse occurred after 12 months, transplant following salvage chemotherapy remains an effective option. If transplant is not the intended pathway, options are not limited to CAR T-cell therapy; bispecific antibodies, antibody-drug conjugates, and monoclonal antibodies represent a genuine third lane of therapy.

Lessons From Three Trials That Didn't Agree

The shift from transplant to CAR T-cell therapy in the second line rests largely on the ZUMA-7 and TRANSFORM trials. ZUMA-7 randomized patients with early relapsed or refractory large B-cell lymphoma to axicabtagene ciloleucel versus standard salvage therapy, with event-free survival as the primary endpoint; the curves separated early, with median event-free survival of roughly eight months versus two months, alongside superior overall response, complete response, and survival rates. TRANSFORM told a similar story with lisocabtagene maraleucel.

BELINDA, evaluating tisagenlecleucel in the same setting, did not show the same separation, but that result reflects factors in the trial's design rather than an indictment of CAR T-cell therapy broadly: patients in BELINDA received more bridging therapy, the product had a longer manufacturing time, and a higher proportion of patients progressed before infusion. The lesson from BELINDA is that logistics matter as much as biology in determining how effective CAR T-cell therapy turns out to be in practice. If a patient with relapsed lymphoma is a CAR T-cell therapy candidate, acting quickly changes outcomes.

Transplant-Ineligible Does Not Mean CAR T-Ineligible

ZUMA-7 and TRANSFORM addressed transplant-eligible patients, but many patients relapsing with DLBCL, including many older adults, are not transplant candidates. The phase 2 PILOT study answered that question directly: transplant-ineligible patients treated with lisocabtagene maraleucel in the second line achieved an overall response rate around 80% and a complete response rate above 50%, with durable responses. That result reframed the conversation: being ineligible for transplant does not mean being ineligible for CAR T-cell therapy.

When a patient is not a transplant candidate, the next useful question is what specifically makes them CAR T-ineligible. Some patients are truly medically ineligible because of frailty, functional status, or organ damage. Others face logistical barriers to CAR T-cell therapy that can potentially be fixed. Still others have temporary barriers, such as an acute infection, an urgent need for cytoreduction, or a transient functional decline related to active disease; for these patients, the relevant question is what can be used to safely bridge them to CAR T-cell therapy, not what can replace it.

Effective Options When CAR T Truly Isn't on the TableEffective Options When CAR T Truly Isn't on the Table

For patients who are genuinely not CAR T candidates, several non-chemotherapy options now have solid supporting data. The phase 3 STARGLO trial compared glofitamab plus gemcitabine and oxaliplatin (GemOx) against rituximab-GemOx in transplant-ineligible relapsed or refractory DLBCL, randomized two-to-one, with overall survival as the primary endpoint. Median overall survival was roughly 25 months versus 12 months, essentially double, with correspondingly better progression-free survival, overall response, and complete response. Cytokine release syndrome (CRS) occurred but was mostly mild, with very few severe cases.

A related approach pairs glofitamab with an antibody-drug conjugate backbone, polatuzumab vedotin, instead of chemotherapy. Compared with rituximab-GemOx, this combination showed better median progression-free survival, around 11 months versus 3 months, along with superior response rates and a safety profile that may be even more favorable: severe CRS occurred in fewer than 5% of patients, and no cases of immune effector cell-associated neurotoxicity syndrome were reported.

Epcoritamab plus GemOx offers a third option in this space, studied as a single-arm trial without a randomized comparator, using the same GemOx backbone as the glofitamab combination. Reported response rates were strong, with an overall response rate around 80% and a complete response rate around 60%. Across these bispecific-based regimens, there is a consistent signal that earlier use tends to produce better responses than later use, a pattern seen with many therapies.

Third Line and Beyond: The Problem Is Choice, Not Scarcity

By the third line and beyond, the challenge is no longer a lack of options but choosing among many. CAR T-cell therapy or bispecific monotherapy remain reasonable where not already used in the second line. Antibody-drug conjugate and multi-target antibody options are also available: polatuzumab-based combinations, such as with rituximab, are effective, though using polatuzumab earlier in treatment may reduce its effectiveness at a later relapse. Loncastuximab tesirine, a CD19-directed antibody-drug conjugate, is effective, with the caveat that CD19 expression should be checked after prior CD19-directed CAR T-cell therapy, since it may be lost. An anti-CD19 monoclonal antibody paired with lenalidomide offers a chemotherapy-free option that many clinicians have found tolerable in practice.

Progression-free survival across these third-line-and-beyond options tends to cluster around four and a half to five months, but that clustering is misleading if read as equivalence: these results come from different trials, different patient populations, different prior therapies, and different lines of use. Bispecific antibodies illustrate this well. For epcoritamab, the overall median progression-free survival is around four months, but among patients who achieve a complete response, median progression-free survival extends to roughly three years. Population-level medians can obscure a genuine tail of durable responders.

Sequencing CAR T-Cell Therapy and Bispecific Antibodies

When a patient is eligible for both CAR T-cell therapy and bispecific antibody therapy, sequencing decisions get more complicated. CAR T-cell therapy currently has the most mature, durable remission data and is a one-time, potentially curative treatment, though the upfront commitment, including cell collection, manufacturing, and associated logistics, is substantial. Bispecific antibodies are off-the-shelf, can start quickly, offer broader feasibility including community-based administration, and carry lower rates of severe CRS and neurotoxicity, though their curative potential remains uncertain, and more mature data on that question will take time.

There is modest data supporting bispecific antibody use after CAR T-cell therapy, and it appears to be efficacious, though not uniformly: patients relapsing early after CAR T-cell therapy, within roughly three months, fare worse than those relapsing after six months or more, and earlier relapse should prompt renewed consideration of transplant. Data on CAR T-cell therapy after prior bispecific antibody exposure raises more caution, since much of it is retrospective and involves small numbers, and mechanistic concerns about T-cell exhaustion and microenvironment changes are reasonable even though the two therapy classes target different antigens (CD19 for CAR T, CD20 for the bispecific antibodies currently in use). A LYSA study using the DESCAR-T registry looked specifically at CD19 CAR T-cell therapy following prior bispecific antibody exposure and found reassuring outcomes, with progression-free and overall survival that compared favorably to bispecific-naive controls after propensity matching, along with similar toxicity.

A Practical Algorithm for Relapsed or Refractory DLBCL

For a patient relapsing after one prior line of therapy, timing again drives the decision. Relapse within 12 months in a CAR T-eligible patient points toward CAR T-cell therapy. Relapse after 12 months in a transplant candidate who proves chemosensitive points toward high-dose chemotherapy followed by autologous transplant. For patients who are not CAR T candidates and not transplant candidates, effective non-chemotherapy options exist, including bispecific-chemotherapy combinations, bispecific-ADC combinations, and other bispecific- or ADC-based regimens, matching the patient's disease biology and functional status to the least burdensome effective option available.

For Patients: Why Your Relapse Timeline Matters

If your lymphoma relapses after initial treatment, one of the first and most important questions your oncologist will ask is when the relapse happened relative to your first treatment. That timing, not just the type of relapse, shapes which treatments make the most sense for you: an early relapse often points toward CAR T-cell therapy, while a later relapse may still make transplant a good option.

You may also hear about bispecific antibodies, a newer class of treatment that does not require the weeks-long preparation that CAR T-cell therapy does and can often be given closer to home. These options are becoming more effective and are being used earlier in treatment as more data become available. If you are found ineligible for one treatment, such as transplant or CAR T-cell therapy, it is worth asking specifically why, since some barriers are temporary or fixable rather than permanent, and effective alternatives exist either way.

Key Takeaways

  • CAR T-cell therapy has replaced salvage chemotherapy and transplant as the preferred second-line option for patients with early relapsed large B-cell lymphoma, based on the ZUMA-7 and TRANSFORM trials.

  • Transplant ineligibility does not mean CAR T ineligibility; the PILOT study showed strong response rates with CAR T-cell therapy in transplant-ineligible patients.

  • For patients who are truly not CAR T candidates, glofitamab- and epcoritamab-based combinations offer durable, largely chemotherapy-free options with strong response rates.

  • Bispecific antibodies can show a low overall median progression-free survival while still producing a meaningful group of durable, long-term responders among patients who achieve complete response.

  • Sequencing CAR T-cell therapy and bispecific antibodies requires individualized judgment; timing of relapse relative to prior treatment, and prior therapy exposure, both influence expected outcomes.

References

  1. Locke FL, Miklos DB, Jacobson CA, et al. Axicabtagene ciloleucel as second-line therapy for large B-cell lymphoma. N Engl J Med. 2022;386:640-654.

  2. Kamdar M, Solomon SR, Arnason J, et al. Lisocabtagene maraleucel versus standard of care with salvage chemotherapy followed by autologous stem cell transplantation as second-line treatment in patients with relapsed or refractory large B-cell lymphoma (TRANSFORM). Lancet. 2022;399:2294-2308.

  3. Sehgal A, Hoda D, Riedell PA, et al. Lisocabtagene maraleucel as second-line therapy in adults with relapsed or refractory large B-cell lymphoma who were not intended for haematopoietic stem cell transplantation (PILOT). Lancet Oncol. 2022;23:1066-1077.

  4. Abramson JS, Ku M, Fox CP, et al. Glofitamab plus gemcitabine and oxaliplatin (GemOx) versus rituximab-GemOx for relapsed or refractory diffuse large B-cell lymphoma (STARGLO). Lancet. 2024;404:1940-1954.

  5. Brody J, Mészáros Joergensen J, Belada D, et al. Epcoritamab plus GemOx in transplant-ineligible relapsed/refractory DLBCL: results from the EPCORE NHL-2 trial. Blood. 2025;145:1621-1633.

  6. Karimi Y, Vose J, Clausen MR, et al. 3-year update from the EPCORE NHL-1 trial. Blood. 2024;144(Suppl 1):4480.

  7. Crochet G, Couturier A, Bachy E, et al. CAR T-cell therapy remains effective in patients with relapsed/refractory B-cell non-Hodgkin lymphoma after bispecific antibody exposure: results of a LYSA study based on the DESCAR-T registry. Blood. 2022;140(Suppl 1):4639-4641.