Lymphoma relapse: don't delay transplant or CAR-T

A 58-year-old man finished six cycles of R-CHOP for a diffuse large B-cell lymphoma seven months ago. The end-of-treatment PET was negative. Now he returns with an axillary lymph node that grew over three weeks, night sweats and an LDH enzyme above normal. The biopsy confirms the relapse. In the outpatient clinic, the plan noted is "do two cycles of salvage chemotherapy and, if he responds, discuss transplant". The referral to the cell therapy center is left for after the second cycle. His wife asks whether there is still a chance of cure. The scene is fictional and was written for teaching.

This text uses the case as a clinical reasoning scenario for students and health professionals. The central question is what to do when a large B-cell lymphoma comes back early, and why the time until referral weighs so heavily. The answer depends on three clues: when the relapse happened, whether the patient can receive intensive treatment and how long the disease allows until the infusion. If you play the case of the day in NEXO, you will recognize the logic: a new clue changes the weight of the others.

Key points

• Relapse within the first 12 months after first-line treatment, or disease that does not respond to it, is a high-risk group with its own rules.
• In this group, two phase 3 studies, ZUMA-7 and TRANSFORM, showed an advantage of CAR-T in second line over salvage chemotherapy with transplant.
• CAR-T requires lymphocyte collection, manufacturing and preparation, and the disease can progress in that interval.
• Those who do not reach the infusion have a very poor outcome, so referral should be made at the time of the relapse biopsy.
• Age and other diseases do not automatically exclude CAR-T; there are data in patients ineligible for transplant.
• Transplant remains a studied option, especially outside the early relapse group.

Why this is a reasoning scenario

Relapse of an aggressive lymphoma is a moment when time works against the patient. The usual reasoning is to give another known treatment and decide later, which seems prudent. But when the next step takes weeks to be ready, postponing the decision can close the door that was meant to stay open.

What the scene teaches

The scene shows a plan in sequence: first salvage, then think. The reasoning calls for asking which group the patient belongs to, which treatment was better in that group in the studies and how long the logistics take. The relapse happened seven months after the end of first line, that is, within 12 months. This clue, which looks like a detail of a date, changes the path.

What is not at stake

The text does not discuss the choice between CAR-T products, the detailed management of toxicities or bispecific antibodies. It also does not cover myeloma, in which CAR-T has other studies and another context, discussed in the text about CAR-T in myeloma. The focus is the initial decision at relapse and the time until the right treatment. Each patient's management follows the service protocol and the team's decision.

What CAR-T is and why time weighs

CAR-T is a therapy made with the patient's own T lymphocytes. They are collected by apheresis, modified in the laboratory to recognize a lymphoma protein, CD19, multiplied and returned by infusion. Before the infusion, the patient receives light chemotherapy to make room for the cells, lymphodepletion.

The dose and the preparation

In the prescribing information for axicabtagene ciloleucel, the target dose is 2 × 10⁶ receptor-positive T cells per kg of body weight, with a maximum of 2 × 10⁸ cells, preceded by lymphodepletion with cyclophosphamide and fludarabine (Yescarta prescribing information, FDA, 2026). The same prescribing information carries a boxed warning for cytokine release syndrome, neurologic toxicity and secondary hematologic malignancies, including T-cell lymphomas. These risks require a prepared center and are part of the conversation with the patient.

Toxicities in practice

Real-world data help size these risks. In the United Kingdom national service, grade 3 or higher cytokine release syndrome occurred in 5 percent of patients and grade 3 or higher neurotoxicity in 18 percent (Kuhnl et al., 2026). In a series of 64 patients treated with two commercial products, 3 percent had these complications at grades 3 and 4 (Schneider et al., 2026). Rates vary by product, patient profile and center experience, and that is why the patient needs to stay near the service in the first weeks after infusion.

The interval between collection and infusion

Between collection and infusion there is manufacturing, transport and preparation. During this period, many patients need treatment to hold the disease, the so-called bridging therapy. In the United Kingdom national CAR-T service, 42 percent of patients needed containment treatment even before collection, and 97 percent received bridging therapy (Kuhnl et al., 2026). The data show that disease that comes back early tends to be aggressive and that every week counts.

Who takes part in the decision

The decision at relapse is not only the hematologist's who follows the patient. The cell therapy center takes part, assessing eligibility and scheduling collection; the apheresis service; the nursing team, which prepares the patient and watches for toxicities; the radiation oncologist, when bridging includes radiation; and the pathologist, who confirms the relapse and the lymphoma type. In the British series, 75 percent of patients were approved for CAR-T within 3 months of the end of first line (Kuhnl et al., 2026), which only happens when this circuit works fast. Each of these professionals can delay or speed up the path, and communication among them is part of the treatment.

The first clue: when the disease came back

The date of relapse defines the group. The two large studies that changed second line included patients with disease refractory to first line or who relapsed within 12 months of it. In the Yescarta prescribing information, this is exactly the second-line indication: large B-cell lymphoma refractory to first-line chemoimmunotherapy or that relapses within 12 months (Yescarta prescribing information, FDA, 2026).

Why this group is different

The classic path at relapse was platinum-based salvage chemotherapy and, for those who responded, high-dose chemotherapy with autologous transplant. In patients with early relapse, this path has a problem: many do not respond to salvage and never reach transplant. In ZUMA-7, of the 179 patients randomized to standard treatment, only 62, or 35 percent, received high doses and transplant, and the most common reason was lack of response to salvage (Yescarta prescribing information, FDA, 2026). In TRANSFORM, 43 of the patients in the standard group, 47 percent, reached transplant (Abramson et al., 2023).

And when the relapse is late

For those who relapse after 12 months and can receive intensive treatment, ZUMA-7 and TRANSFORM do not make the comparison, because these patients were not included. In this scenario, salvage chemotherapy followed by transplant remains the studied path, and CAR-T comes later in the sequence. The date of relapse, therefore, is not an administrative detail: it separates two lines of reasoning.

What ZUMA-7 and TRANSFORM showed

The two studies compared CAR-T with standard treatment in patients who could receive a transplant. The results point in the same direction, but the details teach.

ZUMA-7

ZUMA-7 randomized 180 patients to axicabtagene ciloleucel and 179 to standard treatment. Event-free survival at 18 months was 41.5 percent with CAR-T and 17.0 percent with standard treatment, and the median was 8.3 versus 2.0 months. With a median follow-up of 46.7 months, overall survival was higher with CAR-T, with a hazard ratio of 0.73, even with 57 percent of patients in the standard group receiving cell therapy afterward (Yescarta label, FDA, 2026). The complete response rate was 65 percent versus 32 percent.

TRANSFORM

TRANSFORM randomized 184 patients to lisocabtagene maraleucel or standard treatment. In the primary analysis, the complete response rate was 74 percent versus 43 percent, and median event-free survival was not reached with CAR-T versus 2.4 months with standard treatment (Abramson and colleagues, 2023). At the 3-year follow-up, progression-free survival was 51 percent versus 26.5 percent and overall survival was 63 percent versus 52 percent; 66 percent of patients in the standard group went on to receive CAR-T (Kamdar and colleagues, 2025). The authors describe the treatment as potentially curative in this group.

Why the standard group did so poorly

The low numbers in the standard group do not mean that salvage chemotherapy is a bad treatment in itself. They reflect the type of disease included: a lymphoma that came back early, or that did not respond to first-line treatment at all, tends to resist the next chemotherapy as well. In TRANSFORM, the standard group received three cycles of platinum-based chemoimmunotherapy and, in those who responded, high doses and transplant, and median event-free survival was only 2.4 months (Abramson and colleagues, 2023). CAR-T changes the mechanism of attack: instead of more chemotherapy, it uses cells of the immune system. That is why the advantage appeared precisely in the group in which chemotherapy fails most.

How to read overall survival with crossover

In both studies, many patients in the standard group received CAR-T after failing. This dilutes the difference in overall survival, because part of the control group also receives the new treatment. In TRANSFORM, the hazard ratio for overall survival clearly favored CAR-T when the authors adjusted for this crossover (Kamdar and colleagues, 2025). The lesson for those who read studies is to look at event-free survival and the crossover design before concluding that overall survival "did not change".

Quality of life

In ZUMA-7, quality of life also favored CAR-T. On day 100, the difference on the global health scale was 18.1 points in favor of CAR-T, above the 10-point threshold considered clinically important (Elsawy and colleagues, 2022). For the patient, this means that the more effective treatment also did not weigh more on daily life in the first months.

Table: the path of relapse according to the clues

Situation	Path studied	What weighs in the decision	Main source
Refractory or relapse within 12 months, eligible for transplant	CAR-T in the second line	Refer already at the relapse biopsy	ZUMA-7 and TRANSFORM
Refractory or relapse, ineligible for transplant	CAR-T in the second line is an option	Assess comorbidities with the cell therapy center	ALYCANTE and Breyanzi label
Relapse after 12 months, eligible for transplant	Platinum-based salvage and transplant	Response to salvage	Outside ZUMA-7 and TRANSFORM
Rapidly progressing disease	Containment and bridging therapy	Reach the infusion	United Kingdom series

The second clue: who can receive the treatment

Many patients with relapse are older or have other diseases, and are considered ineligible for transplant. This does not mean, in itself, that they cannot receive CAR-T.

Patients aged 65 or older

In the ZUMA-7 analysis of patients aged 65 or older, 51 received CAR-T and 58 received standard treatment. Median event-free survival was 21.5 months versus 2.5 months, and complete response was 75 percent versus 33 percent (Westin and colleagues, 2023). In the United Kingdom service, the results of patients aged 70 or older were similar to those of younger patients (Kuhnl and colleagues, 2026).

Patients ineligible for transplant

The French ALYCANTE study tested CAR-T in the second line in high-risk patients considered ineligible for transplant due to age 65 or older, a comorbidity index of 3 or more, or a prior transplant. Of the 69 who underwent collection, 62 received the infusion. With a median follow-up of 37.4 months, median progression-free survival was 18.0 months and median overall survival was not reached (Houot et al., 2026). The lisocabtagene maraleucel label includes the indication for patients refractory or in relapse after the first line who are not eligible for transplant due to comorbidities or age (Breyanzi label, FDA, 2026).

What really matters

More than age, what matters is performance status, organ function, active infections and disease control. The Yescarta label advises against giving the product to patients with active infection or inflammatory diseases (Yescarta label, FDA, 2026). This assessment is made by the cell therapy center, and that is why it needs to see the patient early.

The third clue: the clock to infusion

What happens to those who do not reach infusion is the strongest argument for referring early.

Those who do not reach infusion

In the United Kingdom, of 345 patients approved for CAR-T in the second line, 302, or 87.5 percent, received the infusion. The main reason for not receiving it was clinical worsening due to lymphoma progression, in 30 of the 43 patients. Median survival of those who did not reach infusion was only 1.5 months, while 12-month overall survival of those who received it was 73.9 percent (Kuhnl et al., 2026).

Controlling the disease before infusion

A Japanese registry with 913 patients compared two periods and found 6-month progression-free survival of 63.2 percent in the more recent period versus 55.2 percent in the earlier one. The factors associated with the improvement were the use of CAR-T in the second line, better disease control before infusion and a change in product choice (Yagi et al., 2026). In a series of 64 patients in the United States, primary refractory disease and disease in progression at the time of infusion were associated with a worse response (Schneider et al., 2026).

What to do in practice

• Request a biopsy of the relapse and, at the same time, contact the cell therapy center.
• Record the date the first line ended and the date of the relapse.
• Avoid treatments that harm lymphocyte collection without coordinating with the center.
• Plan bridging therapy together with whoever will perform the CAR-T.
• Treat infections and prepare the patient for lymphodepletion.

After the infusion

A PET scan one month after CAR-T helps predict who will relapse. In the British series, patients with partial response and a Deauville score of 4 or 5 had a 6-month risk of progression of 50 and 73 percent, versus 25 percent in those with complete metabolic response (Kuhnl et al., 2026). This allows early planning of the next step for those at high risk.

Risks and reasoning errors

Early relapse is a situation in which small sequencing errors carry a high cost. The most common errors are not of knowledge but of order and timing, as discussed in the text on clinical decision-making.

Waiting for the response to salvage before referring

The plan in the scene leaves the referral until after two cycles. If the disease does not respond, the patient may arrive at the center already in clinical worsening. Taking a diagnostic pause at the time of the biopsy, asking "which group is this patient in?", avoids this delay.

Anchoring on the old routine

For a long time, salvage followed by transplant was the only path. Anchoring on that routine, without looking at the date of the relapse, leads to applying to everyone a plan that, in the early group, lost to CAR-T in two phase 3 studies.

Excluding by age number

Considering the patient too old without a formal assessment excludes people who would benefit, as the data in patients aged 65 or older and ALYCANTE show. The eligibility decision should be shared with the center, as in interprofessional clinical reasoning.

Forgetting the risks

CAR-T has serious toxicities, and the label carries a boxed warning. Ignoring these risks or access to the center is also an error. The conversation with the patient and family should include benefits, risks, logistics and alternatives, with room for the diagnostic uncertainty that still exists in each case.

How NEXO helps train this reasoning

Early relapse is a good example of a clue that seems secondary and changes everything: the date. In NEXO, the case of the day works with progressive clues, and each one asks you to review the hypothesis and the plan. Practicing this move on short cases helps you notice when the calendar is a clinical clue. The method relies on case-based learning with clinical cases.

Three exercises with cases

• Rewrite the scene with the relapse 20 months after the first line and say what changes in the plan and why.
• List the first five actions of the day the relapse biopsy result comes out, in order.
• Write in five sentences the explanation for the family about CAR-T, including benefit, risk and waiting time.

You can also practice the order of hypotheses with the text on differential diagnosis with clues. Those who create cases can contribute through the Creators page.

Frequently asked questions

What is early relapse?

In second-line studies, it is disease that comes back within 12 months after the end of first-line chemoimmunotherapy, or that does not respond to it. It is the group in which ZUMA-7 and TRANSFORM were run.

Did CAR-T replace transplant?

In the early relapse group, CAR-T showed an advantage over salvage with transplant. Outside that group, transplant after salvage remains the path studied in second line.

Why refer before knowing whether salvage works?

Because manufacturing takes time and those who do not reach infusion have a very poor outcome, with median survival of 1.5 months in the United Kingdom series (Kuhnl et al., 2026).

Can an older patient receive CAR-T?

Yes, it depends on the assessment of overall status and comorbidities. There are data on benefit in patients aged 65 or older and in those ineligible for transplant (Houot et al., 2026).

What is bridging therapy?

It is the treatment given between lymphocyte collection and infusion to control the disease. It should be planned with the cell therapy center.

Final checklist

• I recorded the date of the end of first line and the date of relapse.
• I confirmed the relapse by biopsy.
• I contacted the cell therapy center on the same day.
• I assessed overall status, organs and infections.
• I planned bridging therapy with the center.
• I explained to the patient and family the benefits, risks and timing.
• I arranged the PET after infusion and the plan if the response is partial.

Want to practice reasoning with progressive clues? Play the case of the day.

References

1. 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. PMID 36542826.
2. Elsawy M, Chavez JC, Avivi I, et al. Patient-reported outcomes in ZUMA-7, a phase 3 study of axicabtagene ciloleucel in second-line large B-cell lymphoma. Blood. 2022. PMID 35839452.
3. Houot R, Lemoine J, Claudel A, et al. Second-line axicabtagene ciloleucel in large B-cell lymphoma ineligible for transplantation: ALYCANTE final analysis. Hemasphere. 2026. PMID 42774885.
4. Juno Therapeutics, Inc. Breyanzi (lisocabtagene maraleucel), prescribing information approved by the FDA, version of 03/05/2026. Official source.
5. Kamdar M, Solomon SR, Arnason J, et al. Lisocabtagene Maraleucel Versus Standard of Care for Second-Line Relapsed/Refractory Large B-Cell Lymphoma: 3-Year Follow-Up From the Randomized, Phase III TRANSFORM Study. J Clin Oncol. 2025. PMID 40623279.
6. Kite Pharma, Inc. Yescarta (axicabtagene ciloleucel), prescribing information approved by the FDA, version of 08/17/2026. Official source.
7. Kuhnl A, Kirkwood AA, Northend M, et al. Outcomes of second-line axicabtagene ciloleucel for large B-cell lymphoma in the UK. Hemasphere. 2026. PMID 41743266.
8. Kuhnl A, Kirkwood AA, Northend M, et al. Early PET response predicts the risk of relapse after 2L axi-cel in large B-cell lymphoma. Blood Adv. 2026. PMID 41894687.
9. Schneider M, Paruzzo L, Stella F, et al. Naïve CD4(+) T-cells and disease status at CART infusion correlate with clinical outcomes in real-world large B-cell lymphoma patients receiving second-line CAR T therapy. Nat Commun. 2026. PMID 42140929.
10. Westin JR, Locke FL, Dickinson M, et al. Safety and Efficacy of Axicabtagene Ciloleucel versus Standard of Care in Patients 65 Years of Age or Older with Relapsed/Refractory Large B-Cell Lymphoma. Clin Cancer Res. 2023. PMID 36999993.
11. Yagi Y, Kanemasa Y, Terao T, et al. Drivers of temporal improvement in CAR T-cell therapy for large B-cell lymphoma: a Japanese nationwide registry analysis. Blood Adv. 2026. PMID 42018609.