CAR-T Therapy (Cilta-Cel) for Relapsed/Refractory Multiple Myeloma: What the Largest Real-World Study Found

Table of Contents

• Key Points
• Background: Why This Research Matters
• Study Methods: How the Research Was Conducted
• Patient Characteristics: Who Was in the Study
• Key Findings on Safety: Side Effects and Complications
• Key Findings on Response: Did the Cancer Shrink or Disappear?
• Key Findings on Survival: How Long Did Patients Live?
• Clinical Implications: What This Means for Patients
• Limitations: What This Study Could Not Prove
• Recommendations: What Patients Should Know
• Frequently Asked Questions
• Source Information

Key Points

• In 595 real-world patients with relapsed/refractory multiple myeloma, cilta-cel produced an 87% response rate and 73% 12-month progression-free survival.
• Severe cytokine release syndrome occurred in only 4% of patients, though 80% had some degree of CRS.
• Prior BCMA-directed therapy was linked to lower response rates and shorter progression-free survival.
• About 1 in 4 patients had prolonged low blood counts, and nearly half developed infections requiring treatment.
• Patients with heart, lung, or kidney conditions still benefited from cilta-cel, supporting its use in everyday clinical practice.

Background: Why This Research Matters

Multiple myeloma is a blood cancer of plasma cells (cells in the bone marrow that normally make antibodies). When the disease returns or stops responding to standard treatments, it is called relapsed/refractory multiple myeloma (RRMM). For years, patients with RRMM had limited options, especially after several lines of therapy failed.

In February 2022, the U.S. Food and Drug Administration (FDA) approved ciltacabtagene autoleucel (cilta-cel) for patients who had tried at least 4 prior lines of therapy. The approval was later expanded to earlier lines of treatment in April 2024. Cilta-cel is a chimeric antigen receptor T-cell therapy, or CAR-T, a type of living drug made from the patient's own immune cells. Doctors collect T cells (a type of white blood cell), genetically engineer them to recognize and attack cancer cells, and infuse them back into the patient.

Cilta-cel targets a protein called B-cell maturation antigen (BCMA), which is found on the surface of myeloma cells. Each chimeric receptor on the engineered T cells carries two identical camelid-derived heavy-chain antibodies, an unusual design that gives the therapy a strong grip on its target. In the pivotal clinical trial called CARTITUDE-1, cilta-cel produced an overall response rate of 98% and a complete response rate of 83%, with very encouraging survival results. However, clinical trials enroll carefully selected patients who often have fewer health problems than typical patients in the real world.

This study was designed to answer a crucial question: does cilta-cel work as well when given to everyday patients with multiple health conditions and heavily pretreated disease? To find out, researchers turned to the Center for International Blood and Marrow Transplant Research (CIBMTR), a global registry that collects data from more than 375 medical centers on cellular therapies.

Study Methods: How the Research Was Conducted

The researchers analyzed registry data from all patients with RRMM who received commercial (standard-of-care) cilta-cel between March 2022 and December 2023. To be included, patients needed at least 4 prior lines of therapy and a minimum 100-day evaluation after CAR-T infusion. Patients who received a product that did not meet FDA release specifications were excluded.

The study had two primary outcomes: overall response rate (how many patients had their cancer shrink or disappear) and progression-free survival (how long patients lived without the disease getting worse). Secondary outcomes included overall survival (how long patients lived), side effects such as cytokine release syndrome and neurological complications, infections, and treatment-related death.

All data came from the CIBMTR registry, which is a collaboration between the National Marrow Donor Program and the Medical College of Wisconsin. The registry applies automated and manual quality checks, and each contributing center is audited periodically.

Key definitions used in the study:

• High-risk cytogenetics (abnormalities in the cancer cells' chromosomes): presence of t(4;14), t(14;16), t(14;20), and/or deletion 17p at any time before infusion
• Extramedullary disease (EMD): myeloma that has spread outside the bone marrow, detected by imaging (CT, MRI, or PET scans)
• Prolonged cytopenias: failure of neutrophils (infection-fighting white blood cells) to recover above 500/µL, or platelets above 20,000/µL without transfusions, by 30 days after infusion
• Treatment-related mortality (TRM): death occurring without disease progression

A portion of the study period overlapped with a U.S. fludarabine shortage. Fludarabine is normally part of the lymphodepletion chemotherapy (conditioning given before CAR-T to make room for the engineered cells). During the shortage, centers used alternative regimens including bendamustine and cyclophosphamide at their discretion, which allowed a natural comparison of these approaches.

For statistical analysis, researchers used Cox proportional hazards regression models to compare survival outcomes and logistic regression for binary outcomes. They examined 17 variables, including age, sex, race, performance status (a measure of daily function), number of prior therapies, disease stage, tumor burden, laboratory values, and the type of lymphodepletion chemotherapy. Results were considered statistically significant at P values below 0.05, meaning there is less than a 5% chance the finding was due to random variation.

Patient Characteristics: Who Was in the Study

A total of 595 patients with RRMM received standard-of-care cilta-cel. The median age was 64 years (range: 34–84 years), and 57% were male. The majority were White (77%), with Black patients representing 15% and Hispanic patients 9% of the cohort.

Notably, 70% of patients had at least one clinically significant comorbidity (an additional health condition that might have excluded them from clinical trials). Common conditions included:

• Moderate to severe cardiac (heart) disease: 26%
• Pulmonary (lung) disease: 24%
• History of prior malignancies (other cancers): 15%
• Renal (kidney) impairment: 5%
• Cerebrovascular disease (conditions affecting blood flow to the brain): 4%

Most patients (89%) had a good performance status (ECOG 0–1), meaning they were able to carry out daily activities. The study population was heavily pretreated, with a median of 7 prior lines of therapy (range: 4–24). Eighty-five percent had been exposed to three major classes of myeloma drugs (triple-class exposed), and 55% had received five or more different drugs (penta-drug exposed).

High-risk features were common: 27% had high-risk cytogenetics, 40% had advanced (stage II–III) disease by the International Staging System (ISS), 13% had extramedullary disease, and 14% had a high tumor burden, defined as 50% or more plasma cells in the bone marrow.

Eight percent of patients had previously received BCMA-directed therapy, meaning their cancer had already been treated with a drug targeting the same protein that cilta-cel attacks. This included:

• Belantamab mafodotin (an antibody-drug conjugate): 6% (37 patients)
• Teclistamab (a bispecific antibody): 1% (7 patients)
• Prior CAR-T therapy: 0.5% (3 patients)
• More than one BCMA-directed therapy: 1 patient (received both teclistamab and belantamab mafodotin)

Five patients had received a different CAR-T product before, primarily through investigational trials, targeting either BCMA or a different protein called GPRC5D. Fifty-seven percent of patients received bridging therapy (interim treatment to control the disease while waiting for CAR-T manufacturing), and 24% achieved at least a partial response before the infusion. Lymphodepletion consisted of fludarabine plus cyclophosphamide in 78% of patients, bendamustine in 19% (during the fludarabine shortage), and other regimens in 4%.

Key Findings on Safety: Side Effects and Complications

Cytokine release syndrome (CRS)

Cytokine release syndrome is a well-known side effect of CAR-T therapy, caused by a massive release of inflammatory molecules (cytokines) as the engineered T cells multiply and attack cancer cells. Symptoms can include fever, low blood pressure, and oxygen needs.

In this study, CRS of any grade occurred in 80% of patients, but severe cases were uncommon. Grade 2 or higher CRS affected 20%, and grade 3 or higher (serious) CRS occurred in only 4%. The median time to onset was 8 days after infusion. Tragically, there were 4 grade 5 (fatal) CRS events.

On multivariable analysis, patients with a high tumor burden (≥50% bone marrow plasma cells) were substantially more likely to develop grade 2 or higher CRS, with 3.80 times higher odds (95% confidence interval: 2.05–7.03). In practical terms, this means a patient with very heavy marrow involvement has about a 1 in 3 to 1 in 4 chance of moderate-to-severe CRS, compared with roughly 1 in 10 for others — a number worth discussing with the care team before infusion.

Neurological side effects

Immune effector cell-associated neurotoxicity syndrome (ICANS) is a separate complication affecting the brain and nervous system. The median time from infusion to onset was 9 days. Any-grade ICANS occurred in 22% of patients, with grade 2 or higher in 8% and grade 3 or higher in 4%. Two fatal events were observed.

Certain patients faced higher risks of ICANS. On multivariable analysis, the following factors were independently associated with developing any-grade ICANS:

• Age 70 years or older: 1.98 times higher odds (95% CI: 1.17–3.37)
• High plasma cell burden (≥50% in marrow): 2.60 times higher odds (95% CI: 1.35–5.02)
• Hemoglobin below 8 g/dL (significant anemia): 2.54 times higher odds (95% CI: 1.37–4.73)
• Comorbidity index (HCT-CI) of 2 or higher: 1.81 times higher odds (95% CI: 1.20–2.74)

A separate category called non-ICANS neurotoxicity (NINT) was observed in 5% of patients (31 people). This includes delayed or atypical neurological problems that do not fit the classic ICANS pattern. Among these:

• Parkinsonism (symptoms resembling Parkinson's disease, such as stiffness and slow movement): 2.7% (16 patients). Median time to onset was 10 days (range: 8–21).
• Cranial nerve palsies (weakness of nerves in the head and face): 2.5% (15 patients). These most commonly involved cranial nerve VII, which controls facial muscles (12 of 15 patients). Median time to onset was 20 days (range: 17–31).

Because the number of NINT events was small, multivariable analysis could not be performed for this outcome. No deaths were attributed to NINT.

Infections, blood counts, and other complications

Immune effector cell-associated hemophagocytic lymphohistiocytosis-like syndrome (IEC-HS), a rare but serious condition of overwhelming immune activation, occurred in 4% of patients.

Clinically significant infections (infections that required treatment) developed in 47% of patients after CAR-T infusion. Breaking this down:

• Viral infections: 33%
• Bacterial infections: 20%
• Fungal infections: 2% (less common)

Prolonged cytopenias (persistently low blood counts lasting beyond day 30) were observed in 24% of patients. At day 30, neutropenia (low neutrophils) affected 15% of patients, and thrombocytopenia (low platelets) affected 18%.

Second primary malignancies (new, unrelated cancers) occurred in 5% of patients, consisting mostly of nonmelanoma skin cancers. Hematologic (blood) malignancies, including myelodysplastic syndrome (a bone marrow disorder) and non-Hodgkin's lymphoma, occurred in 1%. Importantly, no cancers of T-cell origin were observed — a reassuring finding, since a theoretical risk of CAR-T therapy is that the engineered T cells themselves could turn cancerous.

Deaths and treatment-related mortality

By the last follow-up, 91 patients (15%) had died. Of these deaths, 53 (9%) were due to myeloma progression, 27 (5%) were treatment-related (TRM), and the cause was unknown in 11 patients. The cumulative incidence of TRM at 12 months was 5% (95% CI: 3–7).

The most common cause of treatment-related death was infection, accounting for 11 deaths (including one from COVID-19). Other causes included organ failure (8 deaths), CRS (4), ICANS (2), intracranial hemorrhage or other central nervous system pathology (2), and second primary malignancies, specifically myelodysplastic syndrome (2). Two patients had overlapping attribution to both CRS and organ failure.

Key Findings on Response: Did the Cancer Shrink or Disappear?

The best overall response rate (ORR) — meaning the cancer shrank significantly or disappeared — was 87%. A very good partial response (VGPR) or better was achieved in 75% of patients, and a complete response (CR, meaning no detectable cancer) or better in 35%.

The researchers examined response rates in specific high-risk subgroups, with results shown as ORR and CR rate respectively:

• Patients age 70 or older: 85% ORR, 34% CR
• Patients with ECOG performance status of 2 or higher (more limited daily function): 74% ORR, 32% CR
• Patients with high-risk cytogenetics: 88% ORR, 31% CR
• Patients with extramedullary disease: 77% ORR, 40% CR
• Patients with ISS stage III (advanced disease): 85% ORR, 26% CR
• Patients with prior BCMA-directed therapy: 70% ORR, 17% CR
• Patients who received bendamustine lymphodepletion: 80% ORR, 22% CR

When the researchers adjusted for other factors in multivariable analysis, two characteristics were independently linked to lower complete response rates. Patients who received bendamustine instead of fludarabine/cyclophosphamide for lymphodepletion had roughly half the odds of achieving a complete response (odds ratio 0.50, 95% CI: 0.32–0.78, P = 0.0025). Black or African American patients also had lower odds of complete response compared to White patients (odds ratio 0.58, 95% CI: 0.37–0.92, P = 0.0192) — a finding that requires further investigation into potential biological, access-related, or other contributing factors.

Key Findings on Survival: How Long Did Patients Live?

Survival results were strong. The median progression-free survival (PFS) and overall survival (OS) were not reached after a median follow-up of 12 months (range: 1.1–25.4 months), meaning that more than half of patients had not yet experienced progression or death by the end of the observation period.

The estimated 12-month progression-free survival for the entire cohort was 73% (95% CI: 68–77%), and the 12-month overall survival was 85% (95% CI: 81–88%).

In plain terms: at one year after treatment, roughly 73 out of every 100 patients were alive without their myeloma progressing, and about 85 out of every 100 patients were still alive.

How specific subgroups fared

Extramedullary disease: Patients with EMD had a 12-month PFS of 58%, compared to 72% in patients without EMD (P = 0.08). The 12-month OS was 81% versus 85% (P = 0.60). These differences were noticeable but did not reach statistical significance.

High-risk cytogenetics: Patients with high-risk chromosome abnormalities had a 12-month PFS of 63%, compared to 76% in patients with standard-risk cytogenetics (P = 0.09). The 12-month OS was identical at 84% in both groups (P = 0.91).

Prior BCMA-directed therapy: This was one of the strongest predictors of a worse outcome. Among the 45 patients who had previously received any BCMA-targeted treatment, the 12-month PFS was 51%, compared to 74% in those without prior BCMA exposure (P < 0.001). The difference in 12-month OS was smaller — 81% versus 85% (P = 0.09).

Looking at each type of prior BCMA therapy individually, patients who had received belantamab mafodotin had a 12-month PFS of 50%, and those who had received teclistamab had a 12-month PFS of 54% (P < 0.001 for the comparison). Their 12-month OS estimates were 80% and 86%, respectively (P < 0.0001).

Lymphodepletion regimen during the fludarabine shortage: No significant differences in PFS or OS were observed between patients who received fludarabine/cyclophosphamide versus bendamustine versus other regimens. The 12-month PFS figures were 73%, 66%, and 83% respectively (P = 0.22), and the 12-month OS figures were 84%, 81%, and 100% respectively (P = 0.13).

Disease stage: Patients with ISS stage III disease (the most advanced stage) had notably worse survival. Their 12-month PFS was 50%, compared to 58% for stage II and 79% for stage I patients (P < 0.001). The 12-month OS was 74%, 77%, and 90% in these three groups, respectively (P < 0.001).

Disease status before infusion: Interestingly, how well the disease was controlled before CAR-T infusion did not significantly change survival outcomes. For patients in at least a very good partial response versus partial response versus stable/progressive disease prior to infusion, the 12-month PFS was 75%, 76%, and 72% respectively (P = 0.39), and the 12-month OS was 91%, 89%, and 83% (P = 0.17).

What predicted worse outcomes overall?

Multivariable analysis identified the following characteristics as independently linked to shorter progression-free survival:

• Male sex
• High-risk cytogenetics
• Prior BCMA-directed therapy
• High plasma cell burden (≥50% in bone marrow)
• ECOG performance status of 2 or higher
• Elevated baseline lactate dehydrogenase (LDH), a marker of rapid cell turnover
• Elevated baseline ferritin of 150 ng/mL or higher, a marker of inflammation

For overall survival, the adverse prognostic factors were male sex, ECOG of 2 or higher, elevated LDH, high plasma cell burden, elevated ferritin of 150 ng/mL or higher, and a low platelet count (below 50,000/µL) before infusion.

Clinical Implications: What This Means for Patients

This study provides strong reassurance that cilta-cel works in the real world, not just in carefully selected clinical trial participants. The response rate of 87% and 12-month survival rate of 85% are particularly meaningful given that 70% of the patients had significant comorbidities that may have disqualified them from the CARTITUDE-1 trial.

The results are somewhat lower than those seen in CARTITUDE-1, where the overall response rate was 98%, 12-month PFS was 77%, and OS was 89%. However, this difference likely reflects the sicker, more heavily pretreated population in real-world practice. Notably, the findings are consistent with previously published real-world data showing an ORR of 89%, CR rate of 35%, 12-month PFS of 68%, and OS of 82%.

The study's size allowed researchers to identify patient subgroups that may need closer monitoring or different strategies. The finding that prior BCMA-directed therapy is associated with worse progression-free survival is particularly important, as it has also been reported in patients receiving ide-cel, another CAR-T therapy. With more BCMA- and GPRC5D-targeted immunotherapies becoming available, the researchers note that evidence to guide treatment selection and sequencing is urgently needed. The interval since the most recent BCMA-directed therapy has also been linked to outcomes in other studies, meaning the timing of CAR-T after prior targeted therapy may matter.

One technical note: CIBMTR data reporting requires strict adherence to IMWG response criteria. Patients without urine immunofixation confirmation would not be classified as having a complete response, which may partly explain why the CR rate of 35% appears lower than that reported in some clinical trials.

Limitations: What This Study Could Not Prove

This was an observational registry study, not a randomized controlled trial. That means it can show associations but cannot prove that cilta-cel caused the observed outcomes, and there was no comparison group of patients receiving other treatments.

Follow-up was relatively short, with a median of 12 months. Longer follow-up is needed to assess durability of responses, late side effects, and long-term survival. Data on non-ICANS neurotoxicity relied on registry reporting of specific neurologic symptom patterns; detailed longitudinal neurologic assessments and precise timing of onset were limited by the structure of the CIBMTR forms.

Extramedullary disease classification in the registry includes both true extramedullary and paramedullary disease (disease adjacent to bone), which may have mixed effects on the analysis. Also, a portion of the study coincided with the U.S. fludarabine shortage, so some outcomes data reflect alternative lymphodepletion regimens. Finally, because minority populations made up a relatively small portion of the cohort, the finding about Black or African American race and lower CR rates warrants further study with dedicated analyses.

Recommendations: What Patients Should Know

1. Discuss your full health picture with your care team. Many patients in this study had heart, lung, or kidney conditions and still benefited from cilta-cel. Having a comorbidity does not automatically disqualify you from CAR-T therapy.
2. Ask about prior BCMA-targeted treatments. If you have already received a drug like belantamab mafodotin or teclistamab, be aware that response rates and progression-free survival may be lower. Your doctor can help you weigh whether cilta-cel is the right next step or whether another sequencing strategy might be preferable.
3. Expect close monitoring in the first month. Most CRS begins around day 8 and most ICANS around day 9 after infusion. Parkinsonism symptoms, if they occur, typically begin around day 10, and cranial nerve issues around day 20. Knowing these timelines helps patients and families recognize symptoms early and seek help promptly.
4. Watch for infections. Nearly half of patients developed an infection requiring treatment, and infection was the most common cause of treatment-related death. Report any fever, cough, or other signs of infection to your medical team immediately, even weeks after the infusion.
5. Be patient with blood count recovery. About 1 in 4 patients had prolonged low blood counts lasting past day 30. This may require transfusions or growth factor support, and it typically resolves over time.
6. Do not be discouraged by disease status before infusion. This study found that being in a less-than-perfect remission before CAR-T did not significantly change survival outcomes. Even patients with stable or progressive disease before infusion had similar 12-month PFS and OS to those in deeper remission.
7. Know the signs of late neurological effects. Although rare (affecting about 5% of patients), symptoms like stiffness, slow movement, facial weakness, or other nerve-related problems can appear in the weeks after CAR-T and should be reported promptly.

The researchers concluded that despite advanced disease and a high burden of comorbidities, cilta-cel demonstrated a favorable safety and efficacy profile in standard clinical practice, supporting its continued use. For patients with relapsed/refractory multiple myeloma who have run out of conventional options, this real-world evidence adds confidence that CAR-T therapy is a viable and valuable treatment.

Frequently Asked Questions

Who can receive cilta-cel for multiple myeloma?

Cilta-cel is approved in the U.S. for patients who have tried at least 4 prior lines of therapy. In this large real-world study, most patients had tried a median of 7 prior therapies. Patients with heart, lung, or kidney conditions were included, so having other health issues does not automatically rule out CAR-T therapy.

How well does cilta-cel work in everyday patients?

In a real-world study of 595 patients, 87% responded to treatment, with the cancer shrinking or disappearing. At 12 months, 73% were alive without cancer progression and 85% were still alive. These results are strong, though slightly lower than clinical trial findings, likely because real-world patients were sicker and more heavily pretreated.

Can cilta-cel cause neurological problems?

Yes, some patients develop ICANS, a brain-related complication, usually around day 9. A separate group called non-ICANS neurotoxicity affected 5% of patients, including parkinsonism symptoms around day 10 and cranial nerve palsies like facial weakness around day 20. These rare effects should be reported promptly to your care team.

How long does it take for blood counts to recover after CAR-T therapy?

About 1 in 4 patients had prolonged low blood counts lasting beyond day 30. At day 30, 15% had low neutrophils and 18% had low platelets. This may require transfusions or growth factor support, but it typically resolves over time. Your doctor will monitor blood counts closely during recovery.

Is cilta-cel safe for patients with other health conditions?

In this real-world study, 70% of patients had significant health conditions like heart, lung, or kidney disease, and 89% were still able to carry out daily activities. Despite these comorbidities, 87% responded to treatment. Having a comorbidity does not automatically disqualify you from CAR-T therapy, but your full health picture should be discussed with your care team.

Can a second opinion help me decide if CAR-T therapy (cilta-cel) is right for my relapsed/refractory multiple myeloma, especially if I've already had BCMA-targeted therapy?

In relapsed/refractory multiple myeloma, a second opinion can be useful before choosing CAR-T therapy with cilta-cel. The largest real-world study (595 patients) found that 87% responded overall, but outcomes were worse for patients previously treated with BCMA-targeted drugs such as belantamab mafodotin or teclistamab: their 12-month progression-free survival dropped from 74% to about 51%. Patients with significant heart, lung, or kidney conditions still benefited, so these do not rule out CAR-T. Reviewing whether to proceed with cilta-cel or consider another sequencing strategy is exactly what an independent second opinion can help clarify. Diagnostic Detectives Network provides independent expert second opinions.

Source Information

Original article title: Ciltacabtagene autoleucel for relapsed refractory multiple myeloma - 2026

Authors: Doris K. Hansen, Danai Dima, Hira Mian, Jakob Devos, Ruta Brazauskas, Temitope Oloyede, Aimaz Afrough, Nausheen Ahmed, Larry Anderson, Rahul Banerjee, Jesus G. Berdeja, Aram Bidikian, Binod Dhakal, Ajoy Dias, Yvonne Efebera, Muhammad Salman Faisal, Lohith Gowda, Hamza Hashmi, Abu-Sayeef Mirza, Meera Mohan, Ravi Narra, Ashley E. Rosko, Mark Schroeder, Taiga Nishihori, Heather Landau, Saad Usmani, Marcelo C. Pasquini, Othman S. Akhtar, Surbhi Sidana, and Krina K. Patel

Journal: Blood Cancer Journal (2026). Published by Nature Portfolio. DOI: https://doi.org/10.1038/s41408-026-01496-w

Prior presentation: International Myeloma Society Annual Meeting, Toronto, Canada, September 17–20, 2025.

This patient-friendly article is based on peer-reviewed research. The original manuscript was provided as an unedited, article-in-press version at the time of this writing and may undergo further editing before final publication. It is an open-access article licensed under a Creative Commons Attribution 4.0 International License. Note that this educational summary is not medical advice; patients should discuss their individual situation with their oncology care team.