AUTHORS
Othman Salim Akhtar, MD, Ciara Louise L Freeman, MD, PhD2, Rahul Banerjee, MD3, Kathryn Flynn, PhD4, Aimaz Afrough, MD5, Sabarinath Venniyil Radhakrishnan, MD6, Ravi Kishore Narra, MD.,MS7, Binod Dhakal, MD, MS8, Nadine Abdallah I, MD9, Anita D'Souza Jr, MD10, Marcelo Pasquini, MD11
AFFILIATIONS
1 Medical College of Wisconsin, Milwaukee, Wisconsin, United States
2 Moffitt Cancer Center, Tampa, United States
3 Fred Hutchinson Cancer Center, Seattle, United States
4 Medical College of Wisconsin, Milwaukee, Wisconsin, United States
5 University of Texas Southwestern Medical School, Dallas, Texas, United States
6 Medical College of Wisconsin, Milwaukee, Wisconsin, United States
7 Medical College of Wisconsin, Milwaukee, Wisconsin, United States
8 Medical College of Wisconsin, Milwaukee, Wisconsin, United States
9 Mayo Clinic, Rochester, Minnesota, United States
10 Medical College of Wisconsin, Milwaukee, Wisconsin, United States
11 Cleveland Clinic, Cleveland, United States
ABSTRACT
Background:
Both CAR-T therapy and bispecific antibodies (BsAb) directed against B-cell maturation antigen are approved for patients with RRMM. In the absence of randomized comparative data, treatment selection relies on assessment of the relative risks and benefits of each approach. We sought to quantify how hematologists weigh efficacy, toxicity, and treatment logistics when choosing between CAR-T therapy and BsAb, and to identify clinical scenarios where genuine equipoise exists.
Methods:
We developed a discrete choice experiment (DCE) survey through qualitative physician interviews and pilot testing. In the DCE, a clinical vignette of a patient with triple-class exposed RRMM was presented. Physicians were then asked to choose between two hypothetical treatments for this patient that varied in benefit-risk. Treatment attributes that varied included progression-free survival (PFS), 12-month non-relapse mortality (NRM), grade ≥3 immune-cell effector associated neurotoxicity syndrome (ICANS), delayed neurotoxicity (NT), and treatment modality [one time infusion (CAR-T) versus continuous therapy (BsAb)]. Each respondent completed a total of 16-choice tasks. The survey was distributed nationally to U.S. hematologists/oncologists treating MM. Mixed logit models were used to estimate preference weights and quantify benefit-risk trade-offs. Standardized coefficients (β_std) were generated and reflect the relative influence of each attribute on treatment choice; larger absolute values indicate a greater influence, while the sign indicates the direction of preference.
Results:
We surveyed 180 U.S. - based hematologists/oncologists with a median age of 42 years (IQR, 38-51) and a median of 9 years in practice (IQR, 4-16 years); Eighty-six percent had access to CAR-T therapy and 95% had access to BsAb at their centers. All U.S. regions were represented: Midwest (35%), South (25%), Northeast (20%), and West (19%). Collectively, respondents completed a total of 5,760 choice tasks.
We found that PFS was the dominant driver of treatment choice (βstd, +2.65), followed by 12-month NRM (βstd, -1.05), delayed NT (βstd, -0.54), treatment modality - one time infusion, (βstd. +0.51), and grade ≥3 ICANS (βstd=-0.50). While mixed logit models identified significant heterogeneity in the magnitude of individual preferences (p<0.001), the overall direction remained highly consistent across respondents, with 98.3% favoring longer PFS, 91.1% favoring lower NRM, 87.2% favoring lower delayed NT, 86.5% favoring lower grade ≥3 ICANS, and 85.5% favoring one-time infusion (CAR-T) over continuous therapy (BsAb).
While most respondents preferred CAR-T therapy, treatment preference shifted to BsAb when CAR-T therapy exceeded clearly defined comparative risk thresholds. Respondents chose BsAb when CAR-T was associated with a higher 12-month NRM than the competing BsAb profile (10% vs 5% or 15% vs 10%). The introduction of any delayed NT risk (>0%) strongly favored BsAb, whereas increasing the risk from 3% to 5% had little additional impact on treatment choice (zero-risk bias). Interestingly, grade ≥3 ICANS was comparatively well tolerated, with physicians retaining a preference for CAR-T unless the risk exceeded 10%.
Conclusion
In this national DCE, physicians prioritized PFS, followed by 12-month NRM, when choosing treatment for RRMM, and most preferred CAR-T over BsAb therapy. However, preferences shifted toward BsAb when CAR-T was associated with higher 12-month NRM or any risk of delayed neurotoxicity; the risk of grade ≥3 ICANS did not significantly influence treatment choice. These findings identify the outcomes most likely to drive clinical equipoise and suggest that future comparative studies should focus on patients at elevated risk for NRM or delayed neurotoxicity, for whom treatment preferences are most likely to shift. Further studies are needed to identify patients at increased risk for these outcomes.
DISCLOSURES
Othman Salim Akhtar, MD: No relevant conflicts of interest to declare.