Key Ideas

Early relapse diffuse large B-cell lymphoma (DLBCL), defined as relapse within 12 months of frontline therapy, should prompt evaluation for second-line CAR T-cell therapy rather than default salvage chemotherapy and transplant. Product selection between axicabtagene ciloleucel and lisocabtagene maraleucel, appropriate non-lymphodepleting bridging strategies, and a clear post-CAR-T sequencing plan, including confirming antigen expression before moving to a bispecific antibody, all shape outcomes as much as guideline adherence itself.

A Representative Case: Frontline Treatment Decisions

Consider a 72-year-old man with well-controlled hypertension, stage 3A chronic kidney disease, and a prior deep vein thrombosis, presenting with six weeks of B symptoms, a rapidly enlarging left neck mass, and an LDH twice the upper limit of normal. Imaging and biopsy work-up, PET scan, bone marrow FISH for BCL2/BCL6 to rule out double- or triple-hit lymphoma, echocardiogram given planned cyclophosphamide exposure, and hepatitis/HIV testing, are all appropriate before selecting a frontline regimen, since results shape both regimen choice and CNS risk assessment.

Pathology confirmed DLBCL, not otherwise specified, non-germinal center B-cell (non-GCB) phenotype, without double-hit features. Non-GCB phenotype matters clinically because of constitutive NF-kB signaling, and PET/CT staging showed stage IV disease with extranodal splenic involvement, bulky cervical and retroperitoneal adenopathy, and a high IPI risk score of 4.

The frontline decision came down to standard R-CHOP versus polatuzumab-rituximab-CHP (Pola-R-CHP), per the POLARIX trial. In the 5-year update of POLARIX, Pola-R-CHP maintained a significant PFS benefit over R-CHOP (HR 0.77), though overall survival has not shown a statistically significant difference between arms. In subgroup analysis, Pola-R-CHP appears to work better in non-GCB subtypes, which factored into the decision for this patient. Baseline neuropathy burden is also worth assessing before starting a polatuzumab-based regimen, given its known potential to worsen neuropathy. The key message: this regimen is not oversold as an overall survival advantage, but it earns its place for patients with high IPI scores, extranodal disease, or non-GCB phenotype.

This patient completed six cycles of Pola-R-CHP, complicated by grade 2 neutropenia and grade 1 neuropathy without hospitalization, and achieved a near-complete metabolic response on end-of-treatment PET.

Recognizing and Confirming Early Relapse

Eight months after completing frontline therapy, this patient relapsed with drenching sweats, new abdominal pain, weight loss, and declining performance status, alongside a new FDG-avid mesenteric mass with an SUV max of 19. Biopsy confirmed CD19 and CD20 positivity without double-hit features.

Relapse less than 12 months after frontline therapy makes this patient an ideal candidate for CAR T-cell evaluation. Before referral, confirming CD19 and CD20 positivity on repeat biopsy is an essential first step. Under current guidelines, CAR T-cell therapy is the rule, not the exception, for second-line treatment of primary refractory disease or relapse within 12 months; salvage chemotherapy followed by transplant is now reserved for patients relapsing after 12 months who remain chemosensitive, or for those who are not CAR-T or transplant candidates, for whom bispecific antibodies, clinical trials, or other options become the path forward.

The Evidence Behind Second-Line CAR T-Cell Therapy

Two pivotal trials support second-line CAR T-cell therapy in this setting. ZUMA-7 evaluated axicabtagene ciloleucel (axi-cel) against salvage chemotherapy followed by transplant in early relapsed/refractory DLBCL, demonstrating improved event-free survival, a higher complete response rate, and, in the primary overall survival analysis, a 27.4% reduction in risk of death. Axi-cel carries a comparatively higher rate of ICANS. TRANSFORM evaluated lisocabtagene maraleucel (liso-cel) in a similar population and also showed improved event-free survival versus standard therapy; liso-cel is generally associated with lower rates of early neurotoxicity. A real-world comparison presented at this year's ASCO meeting found similar effectiveness between the two products but a higher signal for early neurotoxicity with axi-cel relative to liso-cel. The BELINDA trial, evaluating tisagenlecleucel in this setting, did not show an event-free survival advantage over standard therapy.

Practical Referral Timeline and Product Selection

A workable community oncology timeline starts with confirming relapse by biopsy (CD19/CD20 positivity) and a baseline infection work-up. Weeks 1-2 focus on apheresis planning and insurance authorization, alongside assessment of disease pace and logistics, distance from a CAR T-cell center matters, since a patient who meets clinical criteria but lives several hours away faces real access barriers.

Bridging therapy should avoid lymphodepletion, since depleting T-cells before collection undermines the eventual product. Reasonable options include short-course steroids for symptomatic patients, involved-site radiation for localized symptomatic disease, and regimens like R-CHOP-based or polatuzumab-based approaches; bendamustine should be avoided as a bridging agent given its lymphodepleting effect.

Product selection between axi-cel and liso-cel is typically made in conjunction with the CAR-T treating center, but general patterns hold: axi-cel tends to suit patients with more rapidly progressive, symptomatic disease, given its strong efficacy, while liso-cel may be preferable for patients without rapidly evolving symptoms and has included a somewhat older patient population in its pivotal trial. Regardless of product, monitoring after infusion should include CRS and ICANS assessment, cytopenias, hypogammaglobulinemia, infection risk with attention to prophylaxis and vaccination status, and caregiver education.

In this case, the patient developed grade 1 CRS on day 7 not requiring tocilizumab, was discharged on day 13, and showed a partial metabolic response on day 30 PET. By day 90, PET showed progressive disease with retained CD20 expression but dim CD19 expression on biopsy.

Sequencing After CAR-T Failure

CD20 retention with dim CD19 expression after CAR-T failure, combined with an ECOG performance status of 2, mild residual neuropathy, and a need for an outpatient-feasible strategy with infection and cytopenia risk in mind, pointed toward a bispecific antibody in this case. Since the patient had two prior lines of therapy and retained CD20 expression, he met the FDA label for a CD20-directed bispecific, and treatment proceeded with standard step-up dosing and CRS monitoring.

Verifying target antigen expression before referring for a bispecific is an essential step, alongside planning step-up monitoring, managing infection risk, and revisiting goals of care at each transition. Where a compelling clinical trial fits the patient's tumor biology, that should always be considered alongside standard-of-care sequencing. Notably, a recent ASCO study comparing CAR T-cell therapy with bispecific antibodies in this setting found that CAR-T produced more durable responses and, in that analysis, a better survival signal, supporting a general preference for CAR-T ahead of bispecific therapy whenever feasible.

When Guidelines Meet Real-World Practice

Clean treatment algorithms rarely map perfectly onto real-world patients. Symptomatic disease, insurance barriers, caregiver availability, and travel distance to a CAR-T center all complicate these pathways in practice, and each deserves consideration before committing to a treatment sequence.

For Patients

Diffuse large B-cell lymphoma, a fast-growing blood cancer, is generally treated first with combination chemotherapy. When the cancer returns within a year of that treatment, current evidence supports moving quickly toward CAR T-cell therapy, which reprograms a patient's own immune cells to attack the cancer, rather than more chemotherapy followed by a stem cell transplant.

Two CAR T-cell products are commonly used, and the right choice often depends on how quickly the disease is progressing, overall health, and which center is providing care. Before starting CAR T-cell therapy, some patients need a short course of treatment to control symptoms while preparations are underway; this bridging treatment is chosen carefully so it doesn't interfere with how well the CAR T-cell product works.

If the lymphoma returns again after CAR T-cell therapy, doctors typically test the tumor to see which markers remain present, since that determines whether a bispecific antibody, another type of immunotherapy, is a good next option. Distance from a treatment center, insurance approval, and support at home all factor into these decisions beyond the clinical guidelines alone. Anyone facing an early relapse should ask their oncologist whether CAR T-cell therapy is appropriate and how quickly that referral should happen.

Key Takeaways

  • Relapse within 12 months of frontline DLBCL therapy should prompt CAR T-cell evaluation as the default second-line strategy under current guidelines, not salvage chemotherapy and transplant.

  • Pola-R-CHP improved PFS over R-CHOP in POLARIX (5-year HR 0.77) without a statistically significant OS benefit; the regimen shows particular benefit in non-GCB, high-IPI, or extranodal disease.

  • ZUMA-7 (axi-cel) and TRANSFORM (liso-cel) both demonstrated improved event-free survival over salvage chemotherapy and transplant in early relapsed/refractory DLBCL; axi-cel carries a higher relative rate of early neurotoxicity.

  • Bridging therapy before CAR-T should avoid lymphodepleting agents like bendamustine; steroids and involved-site radiation are preferred options.

  • After CAR-T failure, confirming CD19/CD20 expression by repeat biopsy is essential before selecting a bispecific antibody, since antigen loss or downregulation is common.

  • A recent ASCO analysis found CAR T-cell therapy produced more durable responses than bispecific antibodies, supporting a general preference for CAR-T when both are options.

  • Real-world barriers, including distance to a CAR-T center, insurance authorization, and caregiver support, meaningfully affect whether guideline-recommended sequencing is actually feasible for a given patient.

References

  1. Locke FL, Miklos DB, Jacobson CA, et al. Axicabtagene Ciloleucel as Second-Line Therapy for Large B-Cell Lymphoma (ZUMA-7). New England Journal of Medicine, 2022; primary overall survival analysis, 2026.

  2. Kamdar M, Solomon SR, Arnason J, et al. Lisocabtagene Maraleucel versus Standard of Care as Second-Line Treatment in Relapsed or Refractory Large B-Cell Lymphoma (TRANSFORM). Lancet, 2022.

  3. Bishop MR, et al. Tisagenlecleucel in Relapsed or Refractory Large B-Cell Lymphoma (BELINDA). New England Journal of Medicine, 2022.

  4. Tilly H, Morschhauser F, Sehn LH, et al. Polatuzumab Vedotin in Previously Untreated Diffuse Large B-Cell Lymphoma (POLARIX); 5-year update, Journal of Clinical Oncology, 2026.

  5. Real-world comparative study of axicabtagene ciloleucel versus lisocabtagene maraleucel in large B-cell lymphoma, effectiveness and neurotoxicity outcomes. American Society of Clinical Oncology, 2026.

  6. Recent ASCO analysis comparing durability of response between CAR T-cell therapy and bispecific antibodies in relapsed/refractory large B-cell lymphoma.