Background

Taxanes are considered first-line chemotherapeutic agents often used in several cancers including those of the breast, ovaries, prostate, gastric, head and neck, and non-small lung cancers [1]. Paclitaxel is one of the important and commonly used antineoplastic agents in this class due to its microtubule-stabilizing mechanism of action. It is commonly used in the management of several cancer types with extensive evidence proving its antimitotic effect [2, 3].

One of the prime adverse effects of taxanes including paclitaxel is peripheral neurotoxicity known as peripheral neuropathy [4]. Up to 97% of patients receiving paclitaxel will develop paclitaxel-induced peripheral neuropathy (PIPN), which turns into a chronic condition in more than 60% of cases [5]. It decreases the efficacy of the chemotherapy by causing patient discomfort and often resulting in dosage reduction or chemotherapy treatment termination. Significantly, those who suffer from chronic neuropathy have considerably poorer long-term quality of life [4].

Peripheral neuropathy is a neurological disorder affecting sensory, motor, and autonomic peripheral nerves [6]. Neuropathic pain resulting from paclitaxel administration develops because of somatosensory nerve damage. Mice administered with paclitaxel exhibit elevated activating transcription factor-3 levels in their large and medium dorsal root ganglion (DRG) neurons, which is indicative of neuronal damage [7]. In PIPN, axonal deterioration and depletion of intra-epidermal nerve fibers (IENFs) was reported, showing that DRG injury is the primary cause of taxane-induced nerve damage [8]. Taxane-induced peripheral neuropathy has been linked to several pathophysiological pathways, including oxidative stress, mitochondrial damage, and microtubule disruption [9]. These pathological pathways are demonstrated in Fig. 1.

Fig. 1
figure 1

Pathophysiology of paclitaxel-induced peripheral neuropathy (PIPN). ATP adenosine triphosphate, C3 complement 3, CX3CL1 C-X3-C motif chemokine ligand 1, IL interleukin, TNF tumor necrosis factor, ROS reactive oxygen species

Full size image

Paclitaxel alters microtubule dynamics which results in impairment in the passage of nutrients, organelles, and neurotransmitters across the neuronal axon leading to axonal degeneration or axonopathy [10]. Also, mitochondrial dysfunction such as morphological alterations, electrolyte imbalance, and reactive oxygen species (ROS) generation has been considered an important element in PIPN. Furthermore, PIPN in rats was linked to the emergence of behaviors caused by pain and mitochondrial disruption in myelinated fibers and C-fibers [11]. Oxidative stress and inflammatory mediators also have a critical importance in the pathophysiology of PIPN [11, 12]. The number of cycles, length of therapy, patient’s age, use of other neurotoxic medications, as well as the presence of risk factors like diabetes, alcoholism, and previous neuropathy, have been linked to the development of PIPN. Additionally, genetic polymorphism in genes like CYP2C8 (cytochrome P450 family 2 subfamily C member 8) and KCNN3 (potassium calcium-activated channel subfamily N member 3) is also linked to the occurrence of PIPN [13].

The most common clinical presentation of PIPN patients is paranesthesia, tingling, and burning “stock and glove.” In more extreme cases, however, it can lead to loss of sensation, motor deficiencies, and autonomic malfunction. Patients typically exhibit sensory symptoms with a “stocking and glove” description, affecting the extremities and extending toward the proximal body parts. Hyperalgesia and allodynia, due to tactile and heat stimuli, may be experienced [14, 15].

Because of the evolving pathophysiologic pathways mechanisms and the diversity of causes and risk factors, preventive interventions are desperately needed to lower the prevalence of PIPN [16]. Despite emerging evidence, there are no established pharmacological interventions for PIPN prevention. This review aims to map the existing literature on interventions evaluated for the prevention of PIPN. A summary and overview of the types of pharmacological interventions studied through preclinical and clinical trials and their effects on outcomes were reported in this review. Additionally, an insight into the effects of genetic polymorphism on the development of PIPN was also reviewed.

Literature search

We conducted a comprehensive search in several databases, including MEDLINE through PubMed, Web of Science, the Cochrane Library, EBSCOhost, and Scopus. Studies that were published in the English language from year 1999 to year 2023 are included in this review. The search keywords used were “prevention,” “neuropathy,” “paclitaxel,” “neurotoxicity,” “chemotherapy-induced peripheral neuropathy,” “controlled clinical trials,” and “preclinical studies.”

Main text

Preclinical studies for the prevention of paclitaxel-induced peripheral neuropathy

Multiple preclinical studies in rodent models indicate that various pharmacologic agents may offer protective effects against peripheral neurotoxicity induced by paclitaxel.

One study in experimental rats showed that the antianginal trimetazidine reduced apoptosis, oxidative stress, and neuroinflammation which are all effects resulting in axonal degeneration and are linked to recurrent paclitaxel administration. The mode of action was shown to be due to the upregulation of Notch1 and progranulin [17]. Additionally, another study further confirmed PIPN reduction by trimetazidine through modulating toll-like receptor 4 (TLR4)/p38 and Klotho protein expression in Swiss albino mice [18].

The angiotensin-II receptor blocker (ARB) losartan demonstrated protective properties against PIPN in experimental rats. Its administration proved its anti-inflammatory effect on microglia in the central nervous system (CNS) through the reduced expression of inflammatory mediators such as tumor necrosis factor-alpha (TNFα) and interleukin‐6 (IL‐6) and stimulation of peroxisome proliferator‐activated receptor gamma (PPARγ) [19]. Similarly, telmisartan, another ARB, reduced PIPN in mice through its inhibition of cytochrome-p450-epoxygenase (CYP2J6) and the prevention of oxidized lipid synthesis [20]. The involvement of Angiotensin-2 was further demonstrated in the preclinical study where ramipril administration attenuated functional neuropathy secondary to paclitaxel in mouse models [21].

An animal study investigating the potential anti-inflammatory role of hesperidin in PIPN has shown positive results. Hesperidin administration reduced oxidative stress and inflammatory response in nerve tissue secondary to paclitaxel use. This is assumed to be because hesperidin’s antioxidant nature allows it to scavenge reactive oxygen species (ROS) generated from mitochondria, subsequently preventing membrane damage caused by ROS [22].

The antioxidant effects of vitamin C and curcumin have been investigated in suppressing PIPN with demonstrated benefits in the reduction of TNFα and IL‐6 levels in the DRG of rats [23, 24]. Melatonin was found to exhibit similar effects in reducing mitochondrial damage and attenuating PIPN in animal studies [25].

Furthermore, rosuvastatin showed a reduction in pro-inflammatory mediators and oxidative stress in paclitaxel-treated mice [26]. This is speculated to be due to possessing pleiotropic and anti-inflammatory properties, further confirming its anti-neuropathic potential in rat models [27].

Medications used in the treatment of neuropathic pain such as duloxetine, pregabalin, and amitriptyline were found to attenuate PIPN in animal models [28,29,30]. The preventive mechanism was found to be via downregulation of pro-inflammatory biomarkers such as IL-6 and TNFα as well as upregulation of the antioxidant capacity.

Metformin was studied in paclitaxel-treated mice, and the results indicate the effectiveness of metformin in attenuating hyperalgesia priming induced by paclitaxel through its stimulation of adenosine monophosphate-activated protein kinase (AMPK) [31].

On the other hand, the effect of alogliptin, a dipeptidyl peptidase 4 (DPP-4) inhibitor, on chemotherapy-induced peripheral neuropathy was investigated using mice, and its protective effect was demonstrated only against oxaliplatin-induced neurotoxicity but not that induced by paclitaxel nor bortezomib [32].

Moreover, co-treatment with the phosphodiesterase (PDE) inhibitor cilostazol halted the dedifferentiation of Schwann cells, secondary to paclitaxel administration, mediated by cyclic adenosine monophosphate (cAMP) signaling and demyelination in a mixed culture of Schwann cells and DRG neurons [33].

The role of glutamate neurotransmitter and the development of PIPN was confirmed in rat models where the co-administration of valproate suppressed glutamate accumulation and paclitaxel-induced mechanical allodynia [34].

It was previously shown that the acute administration of paclitaxel to neuroblastoma cells in culture increased the binding of the cytoplasmic calcium-binding protein neuronal calcium sensor 1 (NCS-1) to the inositol 1,4,5 tris-phosphate receptor (InsP3R) [35, 36]. A preclinical investigation confirmed that a single prophylactic injection of Ibudilast or lithium might inhibit PIPN in mice before they received paclitaxel therapy. These substances work by interfering with the way paclitaxel, NCS-1, and the InsP3R interact [37].

Clinical studies for the prevention of paclitaxel-induced peripheral neuropathy

Several clinical randomized studies have been reported in the literature demonstrating the use of investigational drugs in the prevention of PIPN showing promising evidence of their effectiveness. Table 1 summarizes the clinical studies investigating the potential role of different agents in the prevention of peripheral neuropathy (PN) secondary to paclitaxel chemotherapy.

Table 1 Clinical studies for the prevention of paclitaxel-induced peripheral neuropathy (PIPN)
Full size table

Anticonvulsants

Pregabalin and gabapentin are commonly reported to treat different neuropathic pain. A randomized double-blinded clinical pilot study on the use of pregabalin in preventing PIPN reported no significant difference in worst pain scores between the control arm and the pregabalin arm. Moreover, there were no differences across the arms in the worst, average, and least pain area under the curve (AUCs) throughout the first cycle of therapy (p = 0.48, 0.62, 0.22, and 0.07, respectively) or the maximum of average pain. Additionally, the European Organization of Research and Treatment of Cancer Quality of Life Questionnaire (EORTC-QLQ-CIPN20) sensory subscale did not significantly differ between the two arms according to growth curve models or AUC analysis (p = 0.88 and p = 0.46, respectively) [49].

In another double-blinded, placebo-controlled study 40 breast cancer patients were randomly assigned to receive gabapentin or placebo. In all four cycles, the gabapentin group’s neuropathy was primarily grade 1, with no reported cases of ≥ grade 3 neuropathy. In the gabapentin group, the rate of grade 2 and 3 neuropathy was considerably lower (P < 0.001) than in the placebo group. After four cycles of paclitaxel, the gabapentin group’s Nerve Conduction Velocity (NCV) changed to be lower than that of the placebo group (17.7% vs. 61.0% reduction in NCV for sural nerve and 21.9% vs. 62.5% fall in NCV for peroneal nerve) [38].

Biguanides

A double-blinded randomized controlled trial (RCT) assessing the efficacy of metformin in the prevention of PIPN showed that the development of grade two or more peripheral neuropathy (PN) was significantly lower in the metformin group compared to placebo (p = 0.001). Additionally, the time to develop PN was significantly longer in the metformin group. Furthermore, serum nerve growth factor (NGF) was significantly lower in the metformin group, and comparable levels of serum neurotensin were found in the two study groups [41].

Nutritional supplements

Omega 3 fatty acid protective effect in PIPN was assessed in a double-blinded RCT and showed a significant difference in PN occurrence (OR 0.3, 95% CI (0.10–0.88), p = 0.029). The two study groups did not show a significant trend in terms of PIPN severity differences; nevertheless, the placebo group had greater frequencies of PN in all score categories (0.95% CI (− 2.06 − 0.02), p = 0.054). [40]

Vitamin E neuroprotective effect was evaluated in three RCTs, where the incidence of PIPN and modified peripheral neuropathy (PNP) score were significantly lower with Vitamin E. [50] Also, neurotoxicity was more common in the control group than in the vitamin E-supplemented patients [51]. The frequency of ≥ grade 2 neuropathy, was comparable across the two arms; and only a non-significant difference in grade 3 neuropathy was reported. When comparing the two arms, there was a significant difference in the time length of neuropathy, which was measured from the onset of the first ≥ grade 2 or PN to the point at which it resolves to grade 1 neuropathy [52].

Vitamin B complex when used in conjunction with neurotoxic chemotherapy regimens did not prevent CIPN nor was it more effective than a placebo [58].

One RCT indicated the administration of alpha lipoic acid (ALA) in conjunction with the inhibitor of acetylcholinesterase ipidacrine hydrochloride (IPD) significantly reduced the extent of damage to the sural nerves (SNs) and superficial peroneal nerves (SPNs) caused by paclitaxel in patients prescribed for polychemotherapy (PCT). This was done by identifying significant differences in the electroneuromyography (ENMG) indicators of sensory nerve parameters between the studied groups [55]. Another double-blinded RCT showed that the percent of peripheral neuropathy grade 3 was significantly lower in the ALA group. Furthermore, the FACT-GOG-Ntx-12 questionnaire total score was significantly higher in the ALA group. Regarding the biomarkers, ALA group showed lower levels of brain natriuretic peptides (BNP), tumor necrosis factor-alpha (TNF-α), Malondialdehyde (MDA), and NT in comparison with the control group [62].

A controlled study assessed the oral nutritional supplement Eicosatetraenoic (ONS-EPA) acid effect in patients with advanced non-small cell lung cancer, compared to the control group. Patients in the ONS-EPA group showed significantly reduced neuropathy [56].

Another supplement, acetyl-L-carnitine (ALC) has been studied in PIPN. Over two years, CIPN was statistically significantly worse after twenty-four weeks of ALC therapy [47].

Goshajinkigan (GJG), a traditional Japanese herbal medicine, has been evaluated in a randomized controlled trial showing that (GJG) has a neuroprotective effect and nerve-repairing effect in CIPN [60].

Amino-acids and peptides

A randomized controlled study investigating glutamine in PIPN reported an overall frequency of neuropathy across all grades to be 78% at three months and 80% at six months. At six months, the incidences of grade 1 (48%), grade 2 (22%), and grade 3 (10%) neuropathy were recorded and (20%) did not experience neuropathy. Weekly paclitaxel-induced symptoms were mostly of grades 1 and 2, but not grade 4 symptoms. No significant differences were observed across the treatment groups in terms of the symptoms and weekly PIPN was not improved by glutamine [54].

Glutamate has been evaluated in a double-blinded RCT, showing that the selected dosage regimen was ineffective in the prevention of PIPN. Both groups’ frequency of signs and symptoms was similar; however, the glutamate group’s neurotoxicity symptoms tended to manifest at lower severity levels. Additionally, there was similarity between the two groups in the frequency of aberrant electrodiagnostic findings [45].

A double-blinded RCT evaluated the use of glutathione for the prevention of paclitaxel/carboplatin-induced peripheral neuropathy. The results revealed no statistical significance difference in determining neurotoxicity with grade 2 using the National Cancer Institute Common Terminology Criteria for Adverse Event (NCI-CTCAE) v4.0 scale or in the time to development of neurotoxicity with grade 2. Moreover, no significant difference in neurotoxicity was measured by the EORTC-QLQ-CIPN20 [63].

Cytokines

Another double-blinded RCT showed that Recombinant Human Leukemia Inhibitory Factor (LIF) was not protective against carboplatin/paclitaxel-induced CIPN. There was a comparable standardized composite peripheral neuropathy electrophysiology (CPNE) score between the baseline and cycle 4, the last cycle, or the post-treatment assessment in both groups [57].

Mucolytic agent

N-acetylcysteine was evaluated for its effect on the prevention of PIPN because of its activity in reducing oxidative stress and elimination of ROS. RCT results demonstrated that when compared to the low-dosage group (61.9%) and the control group (100%), the high-dose group’s occurrence of grades 2 to 3 PN was significantly reduced (28.6%). A significant improvement in QOL and modified total neuropathy score (mTNS) scores was recorded. Additionally, there were significant differences in serum MDA levels between the high-dosage and low-dose groups, as well as significantly greater levels of NGF in the high-dose group [42].

Phosphodiesterase inhibitors

Cilostazol has been evaluated for its preventive effect on PIPN, and when comparing the cilostazol group (40%) to the control group (86.7%), there was a significant difference in the occurrence of grade 2 and 3 peripheral neuropathies. The control group had a greater rate of clinically significant worsening in neuropathy-related quality of life compared to the cilostazol group. The cilostazol group showed a greater percent increase in serum NGF from baseline. At the end of the study, the circulation levels of the neurofilament light chain (NfL) were found to be comparable across the two arms [43].

Tetracycline antibiotics

Minocycline was studied in a multicentric, double-blinded, pilot trial and reported no significant dissimilarity during the first cycle of treatment. Still, there was a significant difference in the daily average area under the curve (AUC) pain score attributable to paclitaxel acute pain syndrome (P-APS), in favor of minocycline. Additionally, there was a tendency toward improvement in the daily worst pain AUC score across the 12 cycles. The overall EORTC-QLQ-CIPN20 sensory subscale did not significantly change between minocycline and placebo, despite the decrease in P-APS linked to minocycline usage [48].

Antidepressant

Over a long time, the antidepressant duloxetine has been evaluated for the management of CIPN, and recently, it was assessed in a double-blinded RCT to prevent PIPN. According to Patient Neurotoxicity Questionnaire (PNQ), 10 (50%) of the 20 participants in the placebo group experienced neurotoxicity (two mild cases, three moderate cases, four severe cases, and one disabled case). Nonetheless, two patients in the duloxetine group experienced moderate neurotoxicity. Median motor, sensory latency, and motor velocity were shown to have significant variations across the groups. The relative risk of polyneuropathy (relative risk: 1), however, was comparable between the two groups. According to the findings, an electrodiagnostic investigation supported the possibility that duloxetine could be a beneficial medication for breast cancer patients in reducing PIPN [53].

H2 antagonist

A small, randomized placebo-controlled study was conducted to evaluate lafutidine in the prevention of PIPN. Due to limited recruitment, the planned total of patients was not attained. Neuralgia of grade 2 or above affected 22.2% of the lafutidine group compared to 14.3% in the control group. In the lafutidine group, 100% of the participants had peripheral sensory neuropathy grade 2 or above, compared to 71.4% in the control group (p = 0.175). In neither group, there was evidence of PN of grade 3 or higher. The two groups’ PNQ scores did not differ significantly from one another. Following the fourth cycle, there was a tendency for the lafutidine group to have lower FACT/GOG-Ntx scores than the control group. Between the two groups, there was no statistically significant difference in progression-free survival (PFS) [59].

Others

A randomized double-blinded placebo-controlled study was conducted to assess the effect of ganglioside monosialic acid (GM1) in the prevention of PIPN, which showed that the GM1 group had a lower incidence of PN grade 1 or higher in CTCAE v4.0 grading. Additionally, the GM1 group had a better significant difference in the FACT-NTX score. Moreover, a lower significant difference in the Eastern Cooperative Oncology Group Neuropathy scale (ENS) in the GM1 group was reported [61].

Assessment tools of PIPN including patient-reported outcome measurements (PROMs)

Patient-reported outcome measures (PROMs) are used more widely as a significant tool for the assessment of PIPN and are considered a valuable tool for collecting PN symptoms. PROMs are mostly used as endpoint measures in PIPN treatment and prevention clinical trials, as well as in research settings to characterize the natural history of neuropathy development and recovery. The most investigated PROMs were the EORTC-QLQ-CIPN20 and FACT-GOG-Ntx.

EORTC-QLQ-CIPN20

EORTC-QLQ-CIPN20 is a 12-item quality-of-life questionnaire that was developed to gather information on patients’ experiences with CIPN-related symptoms and functional limitations. The CIPN20 comprises three subscales motor, sensory, and autonomic. Using a four-point rating system, patients indicate how much they have experienced each symptom (or “item”) over the past seven days (1 = Not at all, 2 = A little bit, 3 = Quite a little, and 4 = Very much). Higher scores indicate a greater symptom burden. The three subscales are each calculated as the sum of component items, linearly transformed to a 0–100 scale [64].

FACT-GOG-Ntx

The FACT/GOG-Ntx was developed in cooperation between the Gynecologic Oncology Group (GOG) and the Functional Assessment of Chronic Illness Therapy group. The original 11-item FACT/GOG-Ntx11 questionnaire was created to assess the magnitude of CIPN and its effects on patients’ quality of life in relation to motor, sensory, and auditory neuropathy and dysfunction. A five-point rating system is used for each item (0 = Not at all, 1 = A little bit, 2 = Somewhat, 3 = Quite a bit, and 4 = alot). The items’ scores are reversed according to the FACIT groups’ scoring convention, with greater total scores indicating better quality of life [65, 66].

Genetic single nucleotide polymorphisms associated with PIPN

Polymorphisms associated with paclitaxel metabolism

PIPN sensitivity may be raised by polymorphisms in genes related to the metabolism of the drug. Increased severity of PIPN has often been linked to single nucleotide polymorphisms (SNPs) in the ABCB1 gene [67,68,69]. In previous reports of patients with breast cancer receiving taxanes, SNPs in CYP2C8 and CYP3A4 were identified as causes of ≥ grade 2 CIPN [68, 70, 71].

Polymorphisms associated with microtubule function

Genes linked to microtubule function have been investigated to anticipate their relationship with PIPN since taxanes alter microtubule function and could contribute to PIPN pathogenesis. In 1303 European patients receiving paclitaxel, a SNP in the β tubulin IIb-encoding gene TUBB2A was linked to PIPN [72]. However, in 454 patients with ovarian cancer treated with paclitaxel and carboplatin, additive SNPs in MAPT and GSK3B were linked to reported neuropathy [73].

Polymorphisms associated with inherited neuropathies

The relationship between CIPN and genes linked to hereditary neuropathies has also been investigated. In African American patients receiving paclitaxel, SBF2, linked to Charcot-Marie-Tooth (CMT) disorder, was linked to CIPN [74]. Another study involving 58 individuals receiving paclitaxel discovered that FZD3 was linked to CIPN but not SBF2 [74]. Also, FGD4 was linked to CIPN in a trial investigating 219 breast cancer patients receiving taxanes [75]. In a larger cohort of 855 patients of European origin treated with paclitaxel, findings confirmed variation in FGD4 gene was linked to the reported sensory CIPN [76]. Another trial examining 269 cancer patients undergoing treatment in Alliance N08C1 found that ARHGEF10 was linked to CIPN when 49 CMT genes were examined in blood samples [77]. These results have been proved in 138 patients receiving paclitaxel in Alliance N08CA [78].

Polymorphisms associated with inflammatory pathways

An increasing corpus of research indicates that PIPN is influenced by inflammation, and changes in inflammatory pathways were linked to neuropathic pain [9, 79]. SNP in FCAMR, which encodes the FC receptor, led to a substantial relationship with CIPN in 3,431 breast cancer patients receiving paclitaxel treatment [80].

Polymorphisms associated with ion channels

Moreover, PIPN may potentially result from disruption of neuronal function via ion channels [9, 81] and SCN10A encode sodium channels located in the dorsal root ganglia, specifically Nav1.7 and Nav1.8. Among 186 Japanese patients with taxanes-treated breast and ovarian cancer, a SNP in the encoding gene SCN9A was linked to the development of ≥ grade 2 CIPN [82].

Polymorphisms associated with neuronal function

Genes related to cellular repair pathways and nervous system development and function have been connected to PIPN. In patients receiving taxanes, CIPN is linked to changes in the genes coding the Eph receptors (EPHA4, EPHA5, EPHA6, EPHA8), a class of tyrosine kinase receptors responsible for nerve growth and regulation [83,84,85,86,87]. Also, in 107 patients of gynecologic malignancies undergoing taxane or platinum chemotherapy, polymorphisms in the neural development-related genes SOX10 and GPX7 were linked to CIPN [88].

Future perspectives

Potential difficulties in many clinical trials investigating PIPN prevention include small sample sizes, the absence of placebo control groups, varying dosing and treatment regimens of investigational medications, non-standardized definitions and assessments of neuropathy, and inclusion of cancer patients receiving different chemotherapy treatment protocols. These factors limit the generalizability and interpretation of study findings. To manage these obstacles, future trials should ensure randomized placebo-controlled designs with adequate statistical power. Standardizing objective neuropathy assessments and definitions of clinically significant PIPN across studies would enable cross-trial comparisons. Additionally, narrowing eligibility criteria to specific chemotherapy protocols may yield more homogeneous cohorts for evaluating preventive interventions.

Conclusion

Preventing paclitaxel-induced neuropathy is a complex and evolving field. The literature on PIPN prevention is characterized by the heterogeneity of study designs, patient populations, and outcome measures, making conclusive evidence challenging. The literature suggests a variety of potential approaches such as pharmacological interventions; however, more high-quality research is needed to establish clear recommendations for clinical practice.