Pre-treatment amino acids and risk of paclitaxel-induced peripheral neuropathy in SWOG S0221

• Published in: Cancer chemotherapy and pharmacology 2024-08, Vol.94 (2), p.311-321
• Main Authors: Chen, Ciao-Sin, Zirpoli, Gary, Budd, G. Thomas, Barlow, William E., Pusztai, Lajos, Hortobagyi, Gabriel N., Albain, Kathy S., Godwin, Andrew K., Thompson, Alastair, Henry, N. Lynn, Ambrosone, Christine B., Stringer, Kathleen A., Hertz, Daniel L.
• Format: Article
• Language: English
• Subjects: Adult; Aged; Amino acids; Amino Acids - blood; Antineoplastic Agents, Phytogenic - administration & dosage; Antineoplastic Agents, Phytogenic - adverse effects; Antineoplastic drugs; Body mass index; Breast cancer; Breast Neoplasms - drug therapy; Cancer Research; Chemotherapy; Female; Histidine; Humans; Medicine; Medicine & Public Health; Metabolic pathways; Metabolism; Metabolomics; Middle Aged; Neurotoxicity; Oncology; Original Article; Paclitaxel; Paclitaxel - administration & dosage; Paclitaxel - adverse effects; Peripheral Nervous System Diseases - chemically induced; Peripheral neuropathy; Pharmacology/Toxicology; Regression analysis; Severity of Illness Index; Valine
• ISSN: 0344-5704; 1432-0843; 1432-0843
• DOI: 10.1007/s00280-024-04680-6

Background Chemotherapy-induced peripheral neuropathy (CIPN) is a treatment-limiting and debilitating neurotoxicity of many commonly used anti-cancer agents, including paclitaxel. The objective of this study was to confirm the previously found inverse association between pre-treatment blood concentrations of histidine and CIPN occurrence and examine relationships of other amino acids with CIPN severity. Methods Pre-treatment serum concentrations of 20 amino acids were measured in the SWOG S0221 (NCT00070564) trial of patients with early-stage breast cancer receiving paclitaxel. The associations between amino acids and CIPN severity were tested in regression analysis adjusted for paclitaxel schedule, age, self-reported race, and body mass index with Bonferroni correction. The network of metabolic pathways of amino acids was analyzed using over-representation analysis. The partial correlation network of amino acids was evaluated using a debiased sparse partial correlation algorithm. Results In the primary analysis, histidine concentration was not associated with CIPN occurrence (odds ratio (OR) = 0.97 [0.83, 1.13], p  = 0.72). In secondary analyses, higher concentrations of four amino acids, glutamate (β = 0.58 [0.23, 0.93], p  = 0.001), phenylalanine (β = 0.54 [0.19, 0.89], p  = 0.002), tyrosine (β = 0.57 [0.23, 0.91], p  = 0.001), and valine (β = 0.58 [0.24, 0.92], p  = 0.001) were associated with more severe CIPN, but none of these associations retained significance after adjustment. In the over-representation analysis, no amino acid metabolic pathways were significantly enriched (all FDR > 0.05). In the network of enriched pathways, glutamate metabolism had the highest centrality. Conclusions This analysis showed that pre-treatment serum amino acid concentrations are not strongly predictive of CIPN severity. Prospectively designed studies that assess non-amino acid metabolomics predictors are encouraged.