HFS is a well-documented and relatively common skin reaction that is associated with multiple chemotherapy drugs. Capecitabine, 5-fluorouracil, cytarabine and PLD are the most common drugs that cause HFS [15]. HFS has a variety of symptoms, ranging from mild discomfort to palm and foot pain, which may limit function and hinder the patient's daily activities, but usually disappears within 1 to 5 weeks after stopping the drug [15]. Although HFS is not life-threatening, it may be the main reason for decreased patient compliance and may have a serious impact on quality of life. In addition, with the expansion of the application of PLD in lymphoma, the incidence of HFS may increase. Therefore, the prevention and treatment of these reactions are essential to improve the quality of life of cancer patients and avoid unnecessary dose adjustments that may affect the therapeutic effect.
The mechanism of PLD causing HFS may be related to the strong cytotoxicity inherent to doxorubicin, the long half-life of PLD in the blood, or the interaction between doxorubicin and a large number of Cu(II) ions in skin tissue to produce reactive oxygen species [16]. Due to the rich distribution of capillaries, which are composed of a single layer of endothelial cells, in the fingertips and toes and high blood flow, the drug penetrates from the capillary wall to the interstitial space immediately after slight stimulation; the hands and feet are rich in sweat glands, and the drugs are more permeable to sweat in the stratum corneum [17,18]; moreover, keratinocytes, blood cells and fibroblasts will produce inflammatory cytokines, resulting in vasodilation, increased vascular permeability, redness, fever, and swelling [16]. Eventually, the cumulative effect increases with the toxicity of the chemotherapy cycle, further forming the described HFS lesions. HFS can be alleviated by adjusting the dose or dosing intervals, using cooling methods, wearing loose clothing, and using emollient chemicals. Severe HFS may require delayed chemotherapy.
The early identification of mild symptoms, the improvement of patients' relevant education and the close follow-up of doctors are key elements of HFS prevention management. In a nursing support plan, including instructions for providing information on the potential toxicity of PLD and preventing and treating PLD toxicity led to a very low incidence of severe HFS, not exceeding 4% of patients receiving treatment [19]. The current effective way to control HFS is to modify the treatment method, such as prolonging the dosage interval, reducing the dose or interrupting the use of PLD. The above measures can improve the symptoms of HFS within 1–2 weeks, and for level 1 reactions, supportive treatments and emollient agents can be used. To date, no large-scale controlled study has evaluated the therapeutic effect and preventive measures of HFS.
Regarding the preclinical and clinical pharmacokinetics of PLD, the clearance rate of PLD is easily saturated at higher doses. Related studies have reported that the clearance rate of the standard dose of 40–60 mg/m2 (the dose used by patients with solid tumours) is much slower than that of the lower dose 20 mg/m2 (such as the dose used in Kaposi's sarcoma) [20]. The incidence of HFS is linked to the dose intensity of PLD. When PLD is used at the standard dose of 40 mg·m2, the incidence of HFS is 43.8% (46/105), but when the dose intensity of PLD is 30 mg·m2, the incidence of HFS is only 19.3% (12/62); the results are statistically significant (P = 0.001). Considering the dose dependence, the PLD dose should be individualized.
Animal studies have shown that some doxorubicin is excreted in the urine, but bile is the principal route of doxorubicin excretion after PLD [21]. Consequently, when the patient's liver function is abnormal or when bile excretion is restricted, PLD may accumulate in the body for a long time, thereby causing HFS. This study (Table 4) showed that the occurrence of HFS is related to a history of gallstones (P = 0.025), increased ALT (P = 0.000), increased AST (P = 0.000), and increased GGT (P = 0.000), which are independent risk factors for HFS (Table 5). Hence, in the clinical application of PLD, it is necessary to ask the patient in detail whether he/she has a history of gallstones, to evaluate liver function and other related indicators (ALT, AST, GGT), and then to understand whether the patient has abnormal liver function or abnormal bile excretion. This information provides a reference for avoiding the occurrence of HFS and adjusting the dose of PLD reasonably.
The study found that among the 167 patients with lymphoma who used PLD, 58 patients had HFS, with an incidence of 34.7%. Most of them tolerated PLD well, and most patients with HFS had grade 1–2 HFS, only 4 patients had grade 3 HFS (2.4%), 5 patients postponed the chemotherapy cycle due to HFS of grade 2 or higher, 3 patients adjusted the dose to 75% of the initial dose, and only 1 patient switched to other drugs because of HFS; the HFS of the abovementioned patients recovered well through active symptomatic treatment measures. For any level of response, the incidence of HFS associated with PLD is up to 50%, and for the occurrence of adverse reactions above grade 3, the incidence is approximately 20% [10]. Compared with other tumours, the use of PLD causing HFS in lymphoma is relatively rare, and the severity is greatly reduced. The results of this study (Table 4) show that HL patients are more prone to HFS than NHL patients. Regardless of the lack of a significant difference, it is worthy of further research. Therefore, apart from being related to different tumour types, it may also be because the chemotherapy regimen of NHL patients contains high-dose corticosteroids to avoid allergic reactions on the one hand and prevent and treat PLD-related HFS on the other hand [22].
Compared with conventional preparations, the pharmacokinetic differences among patients treated with PLD are significantly higher [23]. Age, sex, and monocyte count before the cycle seem to be related to the patient's PLD clearance. The PLD clearance rate of young patients (< 60 years old) is approximately twice as fast as that of elderly patients (≥ 60 years old), and the clearance rate of PLD among male patients is faster than that of female patients [24,25]. At the same time, the PLD clearance rate decreased significantly with increasing chemotherapy cycles [26], which is related to the decline in the function of the liver's mononuclear phagocyte system (MPS) [24]. The greater reduction in PLD clearance is also associated with a decrease in monocyte count before the PLD treatment cycle, which indicates that the toxicity of doxorubicin to the MPS (as evidenced by a decrease in peripheral blood monocyte count) reduces the clearance of PLD via the MPS. The reduction in PLD clearance is clinically significant, as a longer PLD half-life has been assessed and found to be associated with a greater risk of skin toxicity [27]. To improve the response of PLD treatment and minimize toxicity, it is necessary to determine the factors related to variability within and between patients using PLD.
In this study, the data summarized in Table 4 reveal that the age of patients with HFS was older than that of patients without HFS, but this difference did not reach statistical significance, and it was inconsistent with the abovementioned applicable theoretical research results. This may be because this study is a retrospective study and not randomized according to age. Most young people with good physical strength are in the group with high drug dose intensity, so they are greatly affected by the dose intensity factor, but the effect of dose intensity on the occurrence of HFS in older patients was also confirmed. Although the incidence rate in women was greater than that in men, the difference was not statistically significant, and the study sample size needs to be further expanded. Unfortunately, we cannot determine the relationship between BMI, monocyte count and the occurrence of HFS from these data.
The present study has several limitations. First, the chemotherapy regimens of the patients in this study were not completely consistent, and the interaction between drugs cannot be ruled out, which may have an impact on the results. Second, this is a retrospective study, and the patients could not be reasonably grouped and stratified; moreover, there may be interference between various factors. These factors should be improved in future research. Notwithstanding its limitations, this study provides references for future large-scale prospective, randomized clinical studies and pharmacokinetic studies. Combining these related factors, further research on the association between PLD pharmacokinetics and clinical outcomes (efficacy and toxicity) as well as subsequent large-scale prospective, randomized clinical studies will lead to the development of strategies to optimize PLD treatment.