Summary

CAR-T cures fewer than half of patients with relapsed large B-cell lymphoma, and outcomes after CAR-T failure depend heavily on when relapse occurs: early relapse carries a dismal prognosis, while late relapse can respond to CD20 bispecific combinations. MRD is promising but not yet actionable outside trials. CAR-T is increasingly best viewed as consolidation after effective debulking, and early relapse after CAR-T calls for a clinical trial.

Key Takeaways

  • CAR-T cures fewer than half of patients with relapsed large B-cell lymphoma, and relapse within 90 days carries a dismal prognosis (fewer than 10% alive long term).

  • Late relapse can respond to CD20 bispecific combinations, while early relapse needs a clinical trial.

  • MRD predicts relapse (persistent MRD at three months relapses about 80% of the time), but it should not drive therapy outside trials.

  • CAR-T should be thought of as consolidation after effective debulking, with PET and MRD monitoring and biopsy to confirm antigen status.

  • In vivo CAR-T and CD19-directed bispecifics are promising, but rare late toxicities may take years to appear.

A Changed Landscape

CAR-T was once seen as a desperate measure, reserved for patients after R-CHOP and platinum-based salvage. That view has slowly changed with approval of several bispecific antibodies and antibody-drug conjugates. When counseling patients before CAR-T, the factors that repeatedly affect long-term success are disease burden, performance status, and organ function. Using CAR-T early, with less tumor burden, leads to better outcomes, so novel therapy combinations are increasingly used before and after apheresis to minimize tumor burden. These newer combinations now give reasonable disease control before cell infusion with less toxicity and, hopefully, better outcomes.

How Poor Are the Outcomes After CAR-T Failure?

CAR-T has had some of the most remarkable results in lymphoma, but long-term follow-up of ZUMA-1 shows that it cures fewer than half of patients with relapsed or refractory disease. When the total cost of the product and the care around it, including IVIG and multiple hospitalizations, runs to half a million dollars or more, the result is about 40% cure. When patients relapse, they do extremely poorly. In a European registry of lymphoma patients who received various salvage treatments after CAR-T, only about one in three or one in four were cured at best.

Timing of Relapse Matters Most

Patients who are refractory to CAR-T or relapse within three months of cell infusion do the worst. In one analysis stratified by time from CAR-T, fewer than 10% of patients who relapsed within the first 90 days were alive long term, while those who relapsed at six months or a year and beyond responded better to salvage and were cured at higher rates.

Barriers in the Clinic

Early post-CAR-T patients often have cytopenias that limit trial candidacy. CD19 loss is less common in high-grade B-cell lymphoma than in leukemia, but it still happens. Performance status is still recovering, and treatment fatigue takes a toll on patients. Patients who spent a month at CAR T center, and hoped to be done are then told they need more treatment, that their local oncologist may not be able to give it, and that they are stuck with the cellular therapy team for an unknown time, with no certainty of cure.

MRD: Promising but Not Yet Actionable

MRD is increasingly integrated as a prognostic marker beyond the PET scan and as a predictor for future intervention, and the lymphoma field is catching up to myeloma. In newly diagnosed patients, cell-free DNA based MRD assay showed extremely good outcomes if MRD cleared after chemotherapy and poor outcomes with persistence. That opens two opportunities: de-escalating therapy in MRD-negative patients and escalating before obvious progression in PET-negative or partial-response patients with persistent MRD.

In TRANSFORM, a phase 3 trial of lisocabtagene maraleucel, blood MRD predicted relapse, but the timing mattered a great deal. Early low-level persistence after infusion can still be cured, but patients who are still MRD positive three months out relapse about 80% of the time. Different MRD technologies exist, and what to do with a positive MRD result outside a trial is still unresolved.

When asked what the practical use of MRD is, the speaker answered that, in practice, it is not yet ready for intervention. Testing is done in the context of trials and consultations, but no one should start a bispecific antibody or other salvage based solely on MRD. That, hopefully, will change.

Salvage Options After CAR-T

Bispecific antibodies and ADCs, alone and in combination, remain very important options. The CAR-T drifts out of the algorithm, and no allogeneic CAR-T for lymphoma is under investigation to the speaker's knowledge.

CD20 Bispecific Antibodies

Three CD20 bispecific agents are FDA approved, among them epcoritamab and glofitamab, and mosunetuzumab. Pivotal trials did not exclude prior CAR-T, but few patients had it, so data after CAR-T are limited, though real-world evidence is emerging from consortia such as CUBIC. In published phase 2 trials, about a third of patients reached a complete response, and combinations work better: bispecific plus ADC or chemotherapy raises response and complete response rates, with almost half of patients reaching complete remission, though in selected trial patients. A CD20 bispecific combination has been the speaker's go-to post-CAR-T salvage for the last year and a half.

CD19-Directed Agents and ADCs

Surovatamig, CD19xCD3 bispecific in a phase 1 trial had a recommended dose of 7.2 mg, and patients with and without prior CAR-T exposure had good outcomes, with 30% to 40% achieving complete response, although follow-up is short. A CD19 bispecific compound from Merck salvaged about 30% of patients after CD19 CAR-T failure.

Loncastuximab tesirine, a CD19 ADC, shows real-world efficacy and can be rapid acting. It needs a CD19 target, and tolerability is the major issue: fluid effusions and edema, with need for high-dose steroid prophylaxis. It is not for everyone, but it can produce durable remission. BV-Lenalidomide plus rituximab is supported by a phase 3 trial, but long-term outcomes in post CAR T setting are unknown.  

CELMoDs and Other New Options

Oral CELMoD agents are under investigation. Their beauty is that they can be combined with other active agents, including chemotherapy and bispecifics, and those trials are among the most exciting. CELMoD consolidation after chemotherapy is also being studied.  

Which Approach Is Best?

The answer depends on timing, CD19 status, comorbidities, and whether the goal is cure or palliation. A recent European registry analysis suggested that bispecific combinations may be the most effective salvage in later relapse, while in early relapse most therapies fail. Even with effective CD20 bispecific or ADC-based therapy, early relapse patients rarely have a durable remission, but in late CAR failure, bispecific combinations can give good results. In selected young patients, consolidative allogeneic transplant is still offered, but only a small minority ever reach deep enough remission to get there.

Where the Field Is Heading

Our view is that R-CHOP has lived its life, and polatuzumab with R-CHOP will probably be replaced by combinations of a CD20 bispecific with chemotherapy or an ADC. CD20 bispecifics will be used before CAR-T in some fashion. MRD-guided intervention is close: a press release on Autolus's ALPHA3, a phase 3 study of an allogeneic CD19 CAR-T given as consolidation to patients who are MRD positive after upfront chemotherapy, reports almost 60% MRD clearance in that high-risk population.

 

Take-Home Messages

CAR-T should be considered a consolidation strategy, not a salvage. Effective debulking agents are available. MRD should be monitored along with the PET scan, and a biopsy should confirm disease and antigen status. For obvious relapse of diffuse large B-cell lymphoma, choose a trial; if none, choose a bispecific antibody-based combination, and for early relapse, a trial is needed.

A Case of Relapse After a CAR T therapy

A patient presented with a large mass causing back pain and biopsy showed high-grade diffuse large B-cell lymphoma. He received Pola-R-CHP for a short duration and then CAR T, achieved a good complete response, and relapsed locally six months later. Repeat Biopsy still showed CD19 and CD20 expression.

Options included glofitamab, and epcoritamab. Loncastuximab tesirine was considered less suitable because of fluid retention, edema, and neuropathy in an older patient, though its rapid action is attractive, and we have used it in salvage and as a bridge after apheresis. Polatuzumab-based therapy was less appealing because POLARIX had no prior CAR-T exposure and the patient had prior polatuzumab exposure. He ultimately received glofitamab plus gemcitabine and oxaliplatin as definitive salvage, achieved another complete response, had cytokine release syndrome managed locally, and remains in complete response.  

Panel Discussion: Sequencing and In Vivo CAR-T

On in vivo CAR-T, the speaker shares the concern: rare biological side effects of CAR T  therapy, such as colitis and parkinsonism, were not known until patients were treated in the real world, so in vivo CAR may produce unexpected adverse events years from now. It is unclear at this time what the long-term impact will be for the patients. 

On sequencing, rechallenging CD19 CAR-T in lymphoma may still give a decent response in some patients, and dual-target CARs may address the problem, though most ongoing trials have excluded prior CAR-T users, so it is hard to say where the sequence will land. On the effect of bispecific antibodies before CAR-T on T-cell fitness, there are no prospective data, and a six-month washout in studies is impractical in practice. The clinician may need supportive therapy to control disease while T cells recover and then collect, or take the chance. The question of frontline CAR-T is about access, since most patients are treated in the community.

For Patients

CAR-T cures fewer than half of people with relapsed large B-cell lymphoma, and when it fails, the timing matters. People whose lymphoma returns within three months do poorly with current treatments, while people who relapse later respond better to antibody treatments that help T cells attack lymphoma, often combined with chemotherapy or another targeted drug.

Doctors are testing blood for tiny amounts of lymphoma DNA, called MRD, which may eventually help decide who needs more treatment, but it is not yet used outside clinical trials to change therapy.

Ask your team how soon after CAR-T your lymphoma returned, whether the tumor still carries the markers CD19 and CD20, and whether a clinical trial is available.

References

  1. Spiegel JY, Dahiya S, Jain MD, et al. Outcomes of patients with large B-cell lymphoma progressing after axicabtagene ciloleucel therapy. Blood. 2021;137(13):1832-1835.

  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(10343):2294-2308.

  3. ASTCT Committee on Practice Guidelines survey on evaluation and management of diffuse large B-cell lymphoma after failure of chimeric antigen receptor T-cell (CAR-T) therapy. Journal details to be confirmed.