Longitudinal neurotoxicity and adaptive dose modification during disitamab vedotin therapy for urothelial carcinoma: A real-world study
Wei Zuo, Chunru Xu, Yu Fan, Kaiwei Yang, Yuke Chen, Yue Mi, Yicong Du, Liqun Zhou, Jian Lin, Qi Tang, Xuesong LiBackground
The maximum tolerated dose paradigm may be suboptimal for antibody-drug conjugates because cumulative toxicities can limit treatment continuity during prolonged therapy.
Objectives
To evaluate longitudinal toxicity patterns, dose-reduction timing, and clinical outcomes in patients with urothelial carcinoma (UC) treated with disitamab vedotin (DV).
Design
Retrospective real-world cohort study.
Methods
A total of 83 patients with histologically confirmed UC who received DV were included. Patients were categorized according to dose-reduction timing as no recorded dose reduction, early reduction within 6 cycles, or late reduction after 6 cycles. Safety, objective response rate (ORR), progression-free survival (PFS), and dose-intensity metrics were evaluated. A 6-month landmark analysis was performed to reduce immortal time bias.
Results
Among 83 patients, 53 had no recorded dose reduction, 6 underwent early reduction, and 24 underwent late reduction. Patients without recorded dose reduction had shorter treatment exposure, suggesting that the absence of dose reduction may partly reflect limited treatment opportunity due to early progression or treatment discontinuation. Peripheral neuropathy was the most common adverse event (33.7%) and occurred more frequently in patients undergoing dose reduction, particularly those with late reduction. Neuropathy-free probability differed significantly among groups (P=0.023). Among 51 patients with measurable disease, tumor response distribution differed according to dose-reduction timing (P=0.033), and PFS differed across groups (P=0.002). Starting dose intensity was associated with ORR (P=0.043). In the 6-month landmark cohort, post-landmark PFS did not differ significantly between patients with and without dose reduction by 6 months (P=0.795).
Conclusion
DV-related neurotoxicity demonstrated a cumulative and time-dependent pattern. Dose reduction was frequently implemented during prolonged treatment and was associated with higher cumulative toxicity exposure. After landmark adjustment, dose reduction was not associated with an apparent loss of clinical benefit, supporting individualized toxicity-guided dose optimization during long-term DV therapy.