Pegylated liposomal doxorubicin-induced hand-foot syndrome predicted by serum metabolomic profiling and prevented by calcium dobesilate

To the Editor: Pegylated liposomal doxorubicin (PLD) has been routinely used to treat patients with various types of cancers, including breast and ovarian cancer. However, it frequently causes hand-foot syndrome (HFS), which always results in a requirement for dose reduction or even drug withdrawal in severe cases.1Reyes-Habito C.M. Roh E.K. Cutaneous reactions to chemotherapeutic drugs and targeted therapy for cancer: part II. Targeted therapy.J Am Acad Dermatol. 2014; 71 (quiz 227-228): 217.e1-217.e11Google Scholar,2Miller K.K. Gorcey L. McLellan B.N. Chemotherapy-induced hand-foot syndrome and nail changes: a review of clinical presentation, etiology, pathogenesis, and management.J Am Acad Dermatol. 2014; 71: 787-794Google Scholar Although some treatments, such as regional cooling, topical urea, antioxidants, and steroids, have the goal of symptom control, there are still no effective treatments.3Bun S. Yunokawa M. Tamaki Y. et al.Symptom management: the utility of regional cooling for hand-foot syndrome induced by pegylated liposomal doxorubicin in ovarian cancer.Support Care Cancer. 2018; 26: 2161-2166Google Scholar Our study objective was to explore the biomarkers that can be used to predict the severity of HFS and to seek effective treatment methods. We used liquid chromatography-mass spectrometry to perform metabolomics profiling of 106 pretreatment serum specimens from breast cancer patients who received neoadjuvant chemotherapy (4 cycles of PLD at 40 mg/m2 and cyclophosphamide at 600 mg/m2, followed by 4 cycles of docetaxel at 90 mg/m2 given every 3 weeks) between March 2017 and August 2018. The patients' HFS severity levels were evaluated according to Common Terminology Criteria for Adverse Events 4.03 criteria. Metabolomic profiles revealed that 43 metabolites were strongly associated with the severity of HFS (Fig 1). A logistic regression model was established by employing 4 metabolites, which could predict the HFS severity before treatment, with a sensitivity of 88.2% and a specificity of 80.8%. In the Sprague Dawley rat model, 30 rats were randomly divided into 3 groups (PLD, PLD + calcium dobesilate [CaD], and vehicle control), with 10 rats per group. Compared with a 60% (6/10) incidence of HFS >2, the administration of CaD greatly reduced the severity of HFS (0/10, no incidence of HFS >2). The pathogenesis was further explored by electron microscopy, Evans blue test and targeted metabonomics. We found that PLD can damage skin capillary endothelial cells, increase their permeability, and make doxorubicin further exudate into skin tissue, leading to skin damage and HFS. In the CaD intervention group, capillary wall injury and doxorubicin exudation were reduced; thus, HFS was relieved. Sixty breast cancer patients were then selected between December 2018 and June 2020. All of the patients were given chemotherapy containing PLD and randomly divided into 2 groups: 30 patients as the control group and 30 patients as the intervention group who were given 500 mg CaD three times a day orally for at least 4 cycles. The incidence of HFS >2 in the CaD intervention group was significantly reduced (10% vs 33.3%, P = .028, Table I). At the same time, there was no increase in adverse events.Table ICharacteristics of breast cancer patients undergoing calcium dobesilate therapy to prevent pegylated liposomal doxorubicin-induced hand-foot syndromeCharacteristicsIntervention group (n = 30)Control group (n = 30)PBody mass index (N, %).605 <2513 (43.3)15 (50.0) ≥2517 (56.7)15 (50.0)Age at diagnosis (N, %)>.99 ≤5018 (60)18 (60.0) >5012 (40)12 (40.0)Tumor size (N, %).105 ≤2 cm4 (13.4)5 (16.7) >2 cm and ≤5 cm25 (83.3)19 (63.3) >5 cm1 (3.3)6 (20)ER (N, %)>.99 Positive19 (63.3)19 (63.3) Negative11 (36.7)11 (36.7)PR (N, %).795 Positive17 (56.7)16 (53.3) Negative13 (43.3)14 (46.7)HER-2 (N, %).194 Positive11 (36.7)16 (53.3) Negative19 (63.3)14 (46.7)Ki-67 (N, %).551 ≤20%5 (16.7)4 (13.3) >20%24 (80)26 (86.7) Uncertain1 (3.3)0 (0)TNM stage (N, %).720 I-II17 (56.7)14 (46.7) III11 (36.7)13 (43.3) IV2 (6.6)3 (10.0)Subtype (N, %).427 ER+/PR+/HER-2−14 (46.6)10 (33.4) HER-2+11 (36.7)16 (53.3) ER−/PR−/HER-2−5 (16.7)4 (13.3)Hand-foot syndrome stage (N, %).028 ≤227 (90)20 (66.7) >23 (10)10 (33.3)AJCC 8th stage of Breast Cancer TNM staging system has been employed.ER, Estrogen receptor; PR, progesterone receptor; TNM, tumor, node, metastasis. Open table in a new tab AJCC 8th stage of Breast Cancer TNM staging system has been employed. ER, Estrogen receptor; PR, progesterone receptor; TNM, tumor, node, metastasis. In this study, metabolic profiles revealed that metabolic biomarkers may be associated with the severity of PLD-induced HFS. A logistic regression model can precisely predict HFS severity before treatment. PLD-induced HFS was related to an increase in local skin capillary permeability, causing the seepage of PLD solution into the skin tissues and skin damage. Furthermore, CaD, an analog of 4-hydroxybenzenesulfonic acid (phenol sulfate), has antioxidative, anti-inflammatory, and antiangiogenic properties and is widely used for the treatment of chronic vascular diseases,4Njau F. Shushakova N. Schenk H. et al.Calcium dobesilate reduces VEGF signaling by interfering with heparan sulfate binding site and protects from vascular complications in diabetic mice.PLoS One. 2020; 15: e0218494Google Scholar,5Voabil P. Liberal J. Leal E.C. et al.Calcium dobesilate is protective against inflammation and oxidative/nitrosative stress in the retina of a type 1 diabetic rat model.Ophthalmic Res. 2017; 58: 150-161Google Scholar and it may play a role in preventing severe HFS. For detailed results, refer to the Supplemental Material (available via Mendeley at https://data.mendeley.com/datasets/5jsw5hcrb9/1). None disclosed.