SUMMARY  

• Johnson & Johnson does not recommend the use of CARVYKTI in a manner that is inconsistent with the approved labeling.
• Parekh et al (2025)¹ presented correlative analyses of peripheral blood and the bone marrow tumor microenvironment (TME) from CARTITUDE‑4 (1-3 prior lines of therapy [LOTs]) and CARTITUDE-1 (≥3 prior LOTs) to evaluate the association of earlier use of CARVYKTI with immune fitness and immune effects.
• Mina et al (2025)² presented an updated treatment-positioning Markov model analysis of the second interim analysis (IA2) of CARTITUDE‑4 and matched Flatiron Health multiple myeloma (MM) electronic health record data to assess the survival benefit of CARVYKTI vs standard of care (SOC) when used as second LOT vs later LOTs for lenalidomide‑refractory MM.

CLINICAL DATA

Correlative Biomarker Analyses From CARTITUDE-1 and CARTITUDE-4 Trials

Parekh et al (2025)¹ presented the correlative analyses of peripheral blood and the bone marrow TME from CARTITUDE‑4 (1-3 prior LOTs) and CARTITUDE-1 (≥3 prior LOTs) to evaluate the association of earlier use of CARVYKTI with immune fitness and immune effects.

Study Design/Methods

• A total of 176 patients in CARTITUDE-4 and 97 patients in CARTITUDE-1 received CARVYKTI as study treatment.
• At the time of screening, bone marrow samples were available for 152 patients from CARTITUDE-4.
  • Day 28 and 6‑month follow‑up bone marrow samples were available for 134 and 133 patients from CARTITUDE‑4, respectively.
  • The RNA bulk sequencing method was used to evaluate the TME in bone marrow aspirate (BMA).
    • Gene signatures and pathways modulated longitudinally were identified using gene set variation analysis and mixed-effects models, with associations examined for progression-free survival (PFS) and prior LOTs.
    • T cell receptor (TCR) repertoire analysis was conducted on BMA from a subset of patients enrolled in CARTITUDE-1.
• At the time of enrollment, peripheral blood samples were available for 84 patients from CARTITUDE-1 and for 164 patients from CARTITUDE-4.
  • Immunophenotyping by flow cytometry was conducted on peripheral blood samples collected from both the CARTITUDE-1 and CARTITUDE-4 studies.
    • Baseline immune fitness was evaluated in patients by prior LOTs and in association with PFS, using nonparametric statistics.

Results

CARTITUDE-4: TME Gene Expression Analysis

Myeloid Cell Activation and Inflammatory Response After CARVYKTI Infusion

• Gene expression analysis of the TME in CARTITUDE-4 patients was performed at baseline (n=142), day 28 after infusion (n=134), and 6 months after infusion (n=133) to evaluate anti-tumor myeloid cell activation and inflammatory state.
• B cells were depleted at the 28-day follow-up and showed partial recovery at the 6-month follow-up, which was accompanied by increased immunoglobulin V(D)J recombination activity.
• Neutrophil degranulation and extravasation were increased at both the 28-day and 6-month follow-ups.
  • The enrichment score for neutrophil degranulation was higher at the 28-day follow-up (adjusted P<0.05) and the 6-month follow-up (adjusted P<0.05) compared to the baseline enrichment score. The difference in the enrichment scores for neutrophil degranulation between the 28-day and 6-month follow-ups was not statistically significant.  
• Tumor-associated macrophage (TAM) levels were increased at the 28-day and 6-month follow-ups.
  • The enrichment score for TAMs was higher at the 28-day follow-up (adjusted P<0.05) and the 6-month follow-up (adjusted P<0.05) compared to the baseline enrichment score. The enrichment score for TAMs was higher at the 28-day follow-up than at the 6-month follow-up (P<0.05).
  • Cytokine signaling pathways driven by M1 macrophages (anti-tumor), such as interleukin (IL)-12, were upregulated, whereas pathways associated with M2 macrophages (pro-tumor), including transforming growth factor beta (TGF-β), were not, indicating that TAMs were likely M1 macrophages.
    • The enrichment score for the IL-12 signal transducer and activator of transcription (STAT) 4 pathway was higher at the 28-day follow-up (adjusted P<0.05) compared to the baseline enrichment score. Differences in the enrichment scores for the IL-12 STAT4 pathway between baseline and 6 months and between day 28 and 6 months were not statistically significant.  
    • Differences in the enrichment score for TGF-β between baseline and 28-day follow-up, baseline and 6-month follow-up, and 28-day and 6-month follow-up were not statistically significant.  

T-Cell Activation and Antigen Presentation After CARVYKTI Infusion

• Gene expression analysis of the TME in CARTITUDE-4 patients was performed at baseline (n=142), day 28 after infusion (n=134), and 6 months after infusion (n=133) to evaluate T cell activation and antigen presentation.
• Activation, differentiation, and proliferation of diverse T-cell subtypes were increased at the 28-day and 6-month follow-ups.
  • The enrichment score for the T-cell receptor alpha chain (TCRA) pathway was higher at the 28-day follow-up (adjusted P<0.05) and the 6-month follow-up (adjusted P<0.05) compared to the baseline enrichment score. The difference in the enrichment scores for the TCRA pathway between the 28-day and 6-month follow-ups was not statistically significant.
  • The enrichment score for the cytotoxic T lymphocyte (CTL) pathway was higher at the 28-day follow-up (adjusted P<0.05) and the 6-month follow-up (adjusted P<0.05) compared to the baseline enrichment score. The enrichment score for the CTL pathway was higher at the 28-day follow-up than at the 6-month follow-up (P<0.05).  
• Antigen presentation capacity was enhanced at both the 28-day and 6-month follow-ups, as indicated by upregulated major histocompatibility complex (MHC)-I and MHC-II pathways.
  • The enrichment score for the MHC pathway was higher at the 28-day follow-up (adjusted P<0.05) and the 6-month follow-up (P<0.05) compared to the baseline enrichment score. The enrichment score for the MHC pathway was higher at the 28-day follow-up than at the 6-month follow-up (adjusted P<0.05).  

TME at Baseline for Patients With 1 or 2 Prior LOTs vs Patients With 3 Prior LOTs

• Patients with 1 or 2 prior LOTs demonstrated a more immunocompetent TME with elevated levels of costimulatory molecules, cluster of differentiation (CD) 4+ T-cell antigen-presenting machinery (MHC-II), and TCRA pathway at baseline compared to those with 3 prior LOTs.
• Gene expression analysis of the TME in CARTITUDE-4 patients was performed at baseline (n=49 for 1 prior LOT; n=57 for 2 prior LOTs; n=36 for 3 prior LOTs) to evaluate differences in enrichment scores for key immune pathways, see Figure: Baseline TME Gene Set Enrichment Scores Grouped by Prior LOT.

Patients With PFS >18 Months vs ≤18 Months

• Gene expression analysis of the TME in CARTITUDE-4 patients with PFS >18 months vs PFS ≤18 months was performed at baseline (n=115 vs n=26), day 28 after infusion (n=115 vs n=18), and 6 months after infusion (n=114 vs n=18).
• No significant differences in relative enrichment to baseline were observed for patients with PFS >18 months at 28-day follow-up and 6-month follow-up for regulatory T cells, epithelial-to-mesenchymal transition (EMT) signature and phosphoinositide 3-kinase class I/protein kinase B (PI3KCI AKT) pathway.
• At 6 months vs baseline patients with shorter PFS (≤18 months) had increased regulatory T cells, suggestive of a more immune-suppressive TME, see Figure: Longitudinal Dynamics Stratified by PFS Group.

Patients With 1 prior LOT vs 3 of Prior LOTs

• Gene expression analysis of the TME in CARTITUDE-4 patients with 1 prior LOT vs 3 prior LOTs was performed at baseline (n=49 vs n=36), day 28 after infusion (n=43 vs n=32), and 6 months after infusion (n=47 vs n=33).
• Patients with more prior LOTs exhibited upregulation of regulatory T cells, EMT signature, and PI3KCI AKT pathway in the TME over time from baseline to day 28 or 6 months, see Figure: Longitudinal Dynamics Stratified by Prior LOT.
• Earlier use of CARVYKTI may support longer PFS (>18 months) given a more immunocompetent TME.

CARTITUDE-1: Endogenous Clonal Expansion of CD4 Memory T Cells and TCR Diversity After CARVYKTI Infusion

• Samples from a subset of CARTITUDE-1 patients (treated at Mount Sinai) were analyzed using TCR sequencing on day 28 after CARVYKTI infusion; patients who attained PFS of ≥5 years had the following characteristics:
  • An increased proportion of expanded clonotypes within CD4 memory T cells was observed from baseline (n=9) to day 28 (n=9) after infusion (P=0.02).
  • Higher TCR repertoire diversity in endogenous T cells was reported in patients achieving PFS ≥5 years (n=5) vs PFS <5 years (n=5) (P=0.03), as observed in CARTITUDE-4 patients with fewer prior LOT, further highlighting the importance of robust TCR activation in driving long-term durable response, see Figure: Swimmer Plot Summarizing PFS and Outcomes.

CARTITUDE-1/CARTITUDE-4 Peripheral Blood Testing

• Patients with high baseline CD4+ naïve T-cell levels had significantly better PFS than those with low baseline CD4+ naïve T-cell levels (P<0.001), see Figure: PFS Stratified by CD4+ Naive T Cells at Baseline.
• Baseline CD4+ naïve T-cell levels were compared across prior LOT groups (n=56 for 1 prior LOT; n=61 for 2 prior LOTs; n=75 for 3-4 prior LOTs; n=54 for ≥5 prior LOTs).
  • Difference in levels between 1 prior LOT vs 2 prior LOTs was not significant (P=0.842).
  • Significant decrease in levels was observed between 1 prior LOT vs 3-4 prior LOTs (P=0.003) and vs ≥5 prior LOTs (P<0.001).
  • Patients with 2 prior LOTs had significantly higher levels compared with 3-4 prior LOTs (P<0.001) and ≥5 prior LOTs (P<0.001).
  • Difference in levels between 3-4 prior LOTs and ≥5 prior LOTs was not significant (P=0.167).

CARVYKTI in Earlier LOTs: Improved Patient Outcomes

• Median PFS was not reached in patients with 1-3 prior LOTs vs 34.9 months in patients with ≥3 prior LOTs.
• Longer overall survival (OS) was observed in patients receiving CARVYKTI in earlier lines of treatment.
• Triple-class exposed patients with 3 prior LOTs from CARTITUDE-1 and CARTITUDE-4 as-treated population achieved a median PFS of 50.4 months.

Treatment-Positioning Model Analysis: cartitude-4 Second Interim Analysis

Mina et al (2025)² presented an updated treatment-positioning Markov model analysis of the IA2 of CARTITUDE‑4 and matched Flatiron Health MM electronic health record data to assess the survival benefit of CARVYKTI vs SOC when used as second LOT vs later LOTs for lenalidomide‑refractory MM.

Study Design/Methods

• Survival outcomes of the following 2 treatment sequences were compared using a Markov model:
  • CARVYKTI in second line (2L) followed by SOC in third or later LOTs (3L+)
  • SOC in 2L followed by CARVYKTI in 3L+
• Inclusion/exclusion criteria used for CARTITUDE-4 were applied to the Flatiron SOC cohort.
• SOC was determined using treatment regimens documented in the Flatiron database for lenalidomide-refractory MM, with variations in therapy distribution between 2L and 3L+ settings.
• Time spent in 2L for CARVYKTI and SOC was based on time to next treatment data in the CARTITUDE-4 and Flatiron 2L subgroups, time in 3L+ was based on OS data from the CARTITUDE-4 and Flatiron 3L+ subgroups.
• Time to next treatment data in the CARTITUDE-4 and Flatiron 2L subgroups were used to determine the time spent in 2L for CARVYKTI and SOC, time in 3L+ was based on OS data from the CARTITUDE-4 and Flatiron 3L+ subgroups.
• Outcomes were focused on median OS and OS rates of 5 and 10 years, respectively.

Results

• At the data cutoff date of May 2024, a total of 208 patients were treated with CARVYKTI in the CARTITUDE-4 intent-to-treat cohort; median follow-up was 33.6 months.
• A total of 1445 observations from the adjusted Flatiron SOC cohort fulfilled the inclusion criteria that were similar to CARTITUDE-4 between January 2020 and May 2024; median follow-up was 23.6 months.
• Key prognostic factors and treatment effect modifiers from CARTITUDE-4 (CARVYKTI) and Flatiron (SOC) were matched to assess outcomes; baseline characteristics are presented in Table: Base Case Patient Characteristics and Model Settings.
• In the Flatiron database, the most commonly used 2L treatments were daratumumab, pomalidomide, and dexamethasone (15.6%); daratumumab, bortezomib, and dexamethasone (11.0%); and other multiple myeloma combinations (15.1%).
• Based on the Markov model that compared data regarding 2L and 3L+ treatment and death, 2L treatment with CARVYKTI resulted in longer OS compared to 3L+ treatment with CARVYKTI after SOC, as shown in Table: OS of Second Interim Analysis by Base Case.
• Outcomes with attrition and distribution (exponential) assumptions are presented in Table: Incremental OS of Using CARVYKTI in 2L vs 3L+ With Alternative Attrition and Distribution Models.

Literature Search

A literature search of MEDLINE^®, Embase^®, BIOSIS Previews^®, and Derwent Drug File databases (and/or other resources, including internal/external databases) was conducted on 04 February 2026.