Why pharmacogenomic biomarkers for chemotherapy-induced peripheral neuropathy fail: a systematic review of genetic associations, replication, and clinical translation.

Introduction

Cancer remains a leading global cause of morbidity and mortality ( Wiernik et al., 2024 ). Despite advances in targeted therapies, traditional chemotherapeutics remain highly effective due to their action on rapidly dividing cells, but it is unclear whether these same mechanisms underlie unintended damage to non-proliferating sensory neurons ( Starobova and Vetter, 2017 ). As a result of recent advances in the early detection of cancer as well as improved survival outcomes ( Zajączkowska et al., 2019 ), the long-term adverse effects of cancer treatment have become increasingly important factors related to an individual’s quality of life ( Glare et al., 2014 ). In some cases, the side effects of cancer treatment can be more disabling than the cancer itself, leading to significant functional impairments that affect both basic and instrumental activities of daily living ( Hekmatnia et al., 2022 ).

Chemotherapy-induced peripheral neuropathy (CIPN) is known as one of the most prevalent and debilitating adverse effects of chemotherapeutic treatments, often manifesting as persistent pain and sensory changes ( Brown et al., 2014 ). Its prevalence varies depending on the agent used, ranging from approximately 19% to over 85% ( Fallon, 2013 ). Specific reported prevalences are particularly high for taxanes (approximately 11%–90%), platinum agents (approximately 19%–90%), vincristine (approximately 30%–90%), and bortezomib (approximately 31%–64%) ( Argyriou et al., 2008 ; Cavaletti and Jakubowiak, 2010 ; Starobova and Vetter, 2017 ; Velasco and Bruna, 2015 ). CIPN presents as a length-dependent sensory neuropathy in a “glove and stocking” pattern, with predominant sensory symptoms causing loss of sensation, gait instability and balance impairment, alongside motor features such as distal weakness ( Jones et al., 2015 ; Park et al., 2017 ). Analgesic treatment remains largely unsatisfactory and to date, there are no clear well established evidence-based therapies for the prevention and management of CIPN ( Sałat, 2020 ).

It is hypothesized that inter-individual differences in CIPN susceptibility may be genetically determined ( Chan et al., 2019 ). Consistent with this hypothesis, a range of genetic factors, including single-nucleotide polymorphisms (SNPs) involved in drug absorption, distribution, metabolism, and excretion have increasingly been recognized as important contributors to this heterogeneity ( Arbitrio et al., 2021 ). As a result, substantial efforts have focused on identifying genetic variants contributing to CIPN risk beyond the direct toxic effects of chemotherapy itself ( Cliff et al., 2017 ; Seretny et al., 2014 ).

Genome-wide association studies (GWAS) provide an unbiased framework for uncovering novel loci in contrast to candidate gene studies which typically focus on identifying genetic variants within biologically plausible drug-response pathways ( Chua et al., 2009 ). Despite these genetic findings, replication across studies remains limited, hindered by small sample sizes, differences in genetic ancestry, and inconsistent CIPN grading methods, to name a few, all of which impede the development of robust and reliable genetic biomarkers ( Hooshmand et al., 2022 ).

A well-recognized gap in this literature has been the underreporting of negative or non-replicated genetic associations. Evidence of publication bias has emerged in CIPN research, with studies reporting significant SNP-CIPN associations more likely to be published than those with null or conflicting results ( Seretny et al., 2014 ). This selective reporting of findings may distort the perceived strength and consistency of genetic effects, inflating effect sizes and obscuring the true biological mechanisms underlying CIPN.

This systematic review provides a comprehensive synthesis of genetic associations with CIPN, including both replicated and non-replicated variants. By integrating positive and negative findings, we evaluate the reliability of reported pharmacogenomic signals, identify sources of publication bias, and highlight priority targets for future investigation. Importantly, we move beyond descriptive synthesis to critically examine why these signals have failed to replicate and translate into clinically actionable biomarkers, underscoring the need for validation in larger, genetically diverse cohorts to enable effective personalised CIPN management.

Methods

This systematic review was conducted and reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta Analyses guidelines (PRISMA) ( Page et al., 2021 ). This study is a systematic review, and we did not conduct a meta-analysis in view of the broad range of factors included. The protocol was registered prospectively with PROSPERO (registration number CRD42025635757).

Search strategy

A comprehensive literature search was conducted (database inception in April 2025) by nine independent reviewers across PubMed, Embase, Cochrane Library, and Web of Science, using MeSH (Medical Subject Headings) and Emtree (Excerpta Medica Thesaurus) terms where applicable ( Figure 1 ). The Systematic Review Accelerator’s Polyglot tool was employed to translate PubMed search terms into equivalent terminology for the other databases ( Clark et al., 2020 ). Detailed search strategy and specific terms are provided in Supplementary Material 1 . In addition, to ensure the completeness of our search, we cross-checked the peripheral neuropathy-related outcomes in the United Kingdom Biobank GWAS Catalog to identify any potentially missed publications ( Sudlow et al., 2015 ). Each study was reviewed by at least two independent reviewers, the review process was managed using Covidence.

Study selection

Studies were included if published in English, involved patients treated with neurotoxic chemotherapies, and examined genetic associations with neuropathy measured using defined assessment tools. Reviews, meta-analyses, animal studies, conference abstracts and case reports were excluded.

A total of 622 records were identified across four databases, with 184 duplicates removed and 438 screened, leading to 111 full-text articles assessed. 50 studies were excluded for unrelated outcomes, duplicate datasets were cross-checked, and screening was managed using Covidence. Ultimately, 72 studies were included in the final review ( Figure 1 ). Of these, 11 studies ( Adjei et al., 2021 ; Chua et al., 2020 ; Kanai et al., 2021 ; Kanai et al., 2016 ; Klumpers et al., 2022 ; Komatsu et al., 2015 ; Mahmoudpour et al., 2018 ; Martin-Guerrero et al., 2019 ; Min et al., 2024 ; Wright et al., 2019 ; Zgheib et al., 2018 ) were included by manual search and are catalogued separately in Supplementary Tables. Following screening and eligibility assessment, the included studies were concentrated in four drug classes: vincristine, taxanes, platinum agents, and bortezomib, reflecting the available published evidence.

Quality assessment

The quality of studies identified using database and register search was evaluated using the Quality of Genetic Association Studies (Q-Genie) tool ( Sohani et al., 2015 ; Sohani et al., 2016 ). Q-Genie was used to assess study quality across key domains including rationale, sample selection, bias, statistics and interpretation, with higher scores indicating better quality. Overall scores ranged from 36 to 67, indicating low to high quality and low to high risk of bias (also see Supplementary Table S1 ).

Data extraction

The following key information from the identified articles was extracted as part of the review: PubMed identifier, first author, publication year, study characteristics, chemotherapeutic agent, measurement tool for phenotype, phenotype trait reported (such as time to onset of CIPN or severity of CIPN), associated single nucleotide polymorphism (SNP) characteristics (p-value reported, effect size, 95% confidence interval). We extracted the effect allele as well as the effect allele frequency if the study authors clearly mentioned a risk or protective allele in their reporting of a SNP.

To maintain consistency in the reporting in cases where an effect allele was not reported for a SNP, we reported the reference and alternate alleles for the rsID denoting the SNP. Information regarding the reference and alternate alleles was obtained using the SNPNexus tool, using GRCh38 as the reference assembly. We also provided information regarding the reported gene traits to provide additional context ( Chelala et al., 2009 ; Dayem Ullah et al., 2012 ; Dayem Ullah et al., 2013 ; Dayem Ullah et al., 2018 ; Oscanoa et al., 2020 ). Information regarding gene attribution and reported gene traits were obtained when available from the NIH NCBI database, including the SNP database (dbSNP) based on the closest gene to the relevant SNP.

To minimize the risk of reporting false positive findings in the discovery cohorts of genome-wide association studies, an upper boundary of uncorrected p < 5 × 10 −5 was set for associated SNPs. In addition, genome-wide significance was defined as the conventional threshold of p < 5 × 10 −8 . For the other study types, a cutoff p < 0.05 was used as the significance threshold.

Formal meta-analysis and statistical tests for publication bias were not undertaken because the included studies lacked sufficient methodological comparability in exposure definitions, phenotyping instruments, and reported effect measures.

Results

We summarize findings from 72 studies grouped by chemotherapeutic agent, with most conducted in predominantly European populations.

Vincristine

15 studies examined genetic associations with vincristine-induced peripheral neuropathy (VIPN), including 4 GWAS ( Broyl et al., 2010 ; Diouf et al., 2015 ; Mufti et al., 2024 ; Yamada et al., 2022 ) and multiple candidate gene analyses ( Aplenc et al., 2003 ; Egbelakin et al., 2011 ; Gutierrez-Camino et al., 2018 ; Sepe et al., 2012 ; Van De Velde et al., 2022 ), with additional non-significant discovery entries retained for completeness ( Cho et al., 2010 ; Christofyllakis et al., 2024 ; Guilhaumou et al., 2011 ; Kishi et al., 2007 ; Moore et al., 2011 ) (for study characteristics see Table 1 ; for full list of studies see Supplementary Material 2 ). Most studies assessed VIPN in pediatric cohorts, predominantly those of European origin diagnosed with acute lymphoblastic leukemia. The mean ages of patients included ranged from about 1 year to 66.5 years ( Table 1 ).

Firstly, CEP72 (involved in chromosome segregation) and ETAA1 (a regulator of replication stress and DNA repair) have been associated with the development of VIPN across GWAS ( Diouf et al., 2015 ) and replicated in an observational pharmacogenetic cohort ( Van De Velde et al., 2022 ) ( Table 2 ). In the discovery GWAS ( Diouf et al., 2015 ), CEP72 rs924607 was associated with increased risk of Grade 2–4 VIPN (odds ratio (OR) = 2.43, 95% CI 1.70–3.49; p = 6.33 × 10 −9 ), while ETAA1 rs17032980 exhibited a similar directional effect (OR = 3.17, 95% CI 1.95–5.17; p = 9.77 × 10 −9 ). In a study undertaken by van de Velde et al. (2022) , the associations were supported by replication studies in which they reported significant effects on a pediatric-modified Total Neuropathy Score (ped-mTNS) for CEP72 rs71585289 (ratio of means = 0.53, 95% CI 0.43–0.66; p < 0.0001) and ETAA1 rs35777125 (ratio of means = 0.36, 95% CI 0.19–0.70; p = 0.0007) ( Table 2 ).

In addition to genomic regions being implicated across studies, GWAS have also identified additional candidate variants ( Supplementary Table S2 ). In Diouf et al. (2015) , MTNR1B rs12786200, within a melatonin receptor gene linked to circadian regulation, glucose metabolism, and potentially neuronal stress responses, was associated with reduced VIPN risk (OR = 0.23, 95% CI 0.13–0.40; p = 1.58 × 10 −7 ), whereas a cluster of TMEM215 SNPs, including rs4463516, showed elevated risk (OR = 2.89, 95% CI 1.90–4.39; p = 6.83 × 10 −7 ). Although the biological function of TMEM215 remains incompletely characterised, transmembrane proteins may influence cellular signalling or membrane homeostasis relevant to neuronal vulnerability.

A more recent GWAS identified multiple loci significantly associated with Grade ≥2 VIPN, including MCM3AP rs1815857 (OR = 6.25, 95% CI 2.91–12.11; p = 3.13 × 10 −8 ), a gene involved in DNA replication and cellular stress responses; SPDYA rs12474420 (OR = 0.25, 95% CI 0.14–0.43; p = 4.40 × 10 −8 ), which regulates cell-cycle progression; METTL8 rs79802223 (OR = 0.17, 95% CI 0.09–0.35; p = 2.46 × 10 −8 ), linked to RNA methylation and post-transcriptional regulation; PDE4D rs12658429 (OR = 0.24, 95% CI 0.14–0.42; p = 4.14 × 10 −8 ), which modulates cyclic AMP signalling and inflammatory pathways; FBN2 rs12656510 (OR = 0.36, 95% CI 0.25–0.53; p = 4.06 × 10 −8 ), associated with extracellular matrix structure; ZFAND3 rs200858088 (OR = 0.07, 95% CI 0.02–0.21; p = 2.53 × 10 −10 ), a zinc-finger protein potentially involved in transcriptional regulation; NFIB rs10961381 (OR = 0.07, 95% CI 0.02–0.24; p = 2.50 × 10 −8 ), a transcription factor implicated in neural development; PAPPA rs12235805 (OR = 0.20, 95% CI 0.11–0.30; p = 3.04 × 10 −8 ), which regulates insulin-like growth factor bioavailability; LRRTM3 rs10997459 (OR = 0.14, 95% CI 0.06–0.32; p = 4.84 × 10 −8 ), a neuronal synaptic organiser; and NRG3 rs12253008 (OR = 0.15, 95% CI 0.07–0.33; p = 3.19 × 10 −8 ), a member of the neuregulin family involved in neuronal development and Schwann cell signalling. These GWAS-identified loci have not yet been replicated in independent cohorts, underscoring the need for further validation.

Several candidate gene studies have also reported associations between specific variants and VIPN, spanning multiple functional categories ( Supplementary Table S3 ). In a prospective pharmacogenetic study, Van De Velde et al. (2022) identified NDRG1 rs2272653 as protective (OR = 0.49, 95% CI 0.40–0.60; p < 0.0001), consistent with the known role of NDRG1 in myelin maintenance and hereditary neuropathy; GARS rs1049402 (OR = 0.47, 95% CI 0.34–0.64; p = 0.0013), relevant because mutations in GARS cause axonal Charcot–Marie–Tooth disease; FGD4 rs12823621 (OR = 1.88, 95% CI 1.36–2.59; p < 0.0001), another Charcot–Marie–Tooth gene involved in Schwann cell and cytoskeletal function; and SEPTIN9 rs11650934 (OR = 0.81, 95% CI 0.48–1.38; p < 0.0001), which participates in cytoskeletal organisation and vesicle trafficking ( Terrazzino et al., 2015 ). In a separate study, Gutierrez-Camino et al. (2018) reported that miR-3117 rs12402181 was protective (ratio of means = 0.16, 95% CI 0.05–0.55; adjusted p = 0.0026), whereas miR-4481 rs7896283 increased risk (OR = 3.02, 95% CI 1.37–6.69; p = 0.0103) for overall VIPN severity ( Gutierrez-Camino et al., 2018 ). These microRNA loci may influence VIPN susceptibility through post-transcriptional regulation of genes involved in neuronal repair, inflammation, or axonal homeostasis.

Several other discovery variants reported across multiple studies did not reach statistical significance or have not been independently replicated ( Broyl et al., 2010 ; Cho et al., 2010 ; Christofyllakis et al., 2024 ; Guilhaumou et al., 2011 ; Kishi et al., 2007 ; Moore et al., 2011 ; Yamada et al., 2022 ) and are catalogued in Supplementary Table S4 .

Taken together, the most consistently replicated signals include CEP72 (chromosome segregation) and ETAA1 (replication stress response) from GWAS, as well as candidate gene associations involving axonal/neuronal structure (NDRG1, GARS, FGD4), miRNA regulation (miR-3117, miR-4481), and drug metabolism (CYP3A4/5).

Taxanes

24 studies investigated the association between genetic variants and taxane-induced peripheral neuropathy (TIPN), including 4 GWAS ( Baldwin et al., 2012 ; Hooshmand et al., 2022 ; Leandro-García et al., 2013 ; Sucheston-Campbell et al., 2018 ) and multiple candidate gene studies across diverse cohorts (see Table 3 for study characteristics, for full list of studies see Supplementary Material 2 ). Most studies were conducted in adults with breast cancer, although some included older patients (≥65 years) and individuals treated with taxanes for other solid tumors ( Table 3 ). The majority of cohorts were of European ancestry, with few studies conducted in Asian ancestry populations ( Nakayama et al., 2024 ; Tanabe et al., 2020 ). Reported patient ages ranged from 18 years ( De Graan et al., 2013 ) to 87 years ( Leskelä et al., 2011 ), and sample sizes varied widely, from 25 participants ( Sissung et al., 2006 ) to 1,303 ( Abraham et al., 2014 ).

A consistent signal of suggestive significance emerged for EPHA5 (rs7349683) , an ephrin receptor gene implicated in axon guidance, which was associated with increased TIPN risk in both GWAS and candidate gene studies (p = 9.6 × 10 −7 ; p = 0.03, respectively) ( Baldwin et al., 2012 ; Boora et al., 2015 ) ( Table 4 ). Baldwin et al. (2012) also identified FGD4 ( rs10771973 ; OR = 1.57, p = 2.6 × 10 −6 ) alongside EPHA5 ( rs7349683 ; OR = 1.63, p = 9.6 × 10 −7 ) as suggestively significant loci with known roles in axonal guidance ( Supplementary Table S5 ) ( Baldwin et al., 2012 ). Similarly, Leandro-García et al. (2013) reported EPHA6 rs301927 as a risk allele (OR = 2.35, p = 3.44 × 10 −5 ), suggesting a potential neuronal targeting mechanism ( Supplementary Table S6 ) ( Leandro-García et al., 2013 ).

Another significant and replicated finding was rs7833751 in FZD3 , a gene involved in Wnt signaling and neurite development, which was associated with reduced TIPN risk (p = 0.0011 and 7.5 × 10 −9 ) ( Baldwin et al., 2012 ; Chen et al., 2020 ) ( Table 2 ). SCN9A , encoding a voltage-gated sodium channel expressed in nociceptive neurons, showed consistent associations across breast cancer cohorts in candidate gene studies ( rs13017637 ; OR = 3.46, 95% CI 1.46–8.24, p < 0.005) ( Supplementary Table S6 ) ( Tanabe et al., 2020 ). Another plausible candidate was S1PR1 ( rs74497159) , which showed suggestively significant association with the risk of developing Grade 2 or higher paclitaxel-induced peripheral neuropathy in a GWAS meta-analysis (p = 3.62 × 10 −7 ) ( Supplementary Table S11 ) ( Chua et al., 2020 ).

Across the reviewed studies, a number of key variants in genes related to taxane metabolism and transport were identified ( Supplementary Tables S5, S6 ). For example, variants in CYP2C8 which is a key taxane-metabolizing enzyme, were variably associated with neuropathy risk. CYP2C8 rs1058930 increased susceptibility in some studies (hazards ratio (HR) = 1.38, 95% CI 1.03–1.86) ( Abraham et al., 2014 ) but showed inconsistent directionality across other studies (HR range = 0.53–1.46) ( Leskelä et al., 2011 ). Likewise, CYP3A5 rs776746 was linked to decreased TIPN risk (HR = 0.51), likely reflecting enhanced taxane clearance ( Leskelä et al., 2011 ). CYP1B1 rs1056836 , involved in xenobiotic and estrogen metabolism, conferred modest protection (p = 0.01–0.02; HR = 0.83; OR = 0.81) ( Abraham et al., 2014 ), though replication success varied ( Boora et al., 2015 ; Rizzo et al., 2010 ).

Transporter genes also featured prominently in the studies that were included in the review ( Supplementary Table S6 ). ABCB1 variants, including rs3213619 and rs1128503 , were associated with altered neuropathy risk, showing consistent or increased effects across independent cohorts ( Abraham et al., 2014 ; Boora et al., 2015 ; Vargas-Aliaga et al., 2024 ). ABCC2 rs8187710 and SLCO1B1 rs3829306 were associated with reduced TIPN risk (p = 0.02; HR = 0.67–0.71), consistent with their roles in systemic drug clearance ( Abraham et al., 2014 ). Conversely, TUBB2A rs9501929 was linked to increased risk (p = 0.005; HR = 1.60), implicating microtubule structural variability in taxane-specific neurotoxicity ( Abraham et al., 2014 ).

Not all the genetic findings were reproducible, and further validation is needed as part of future interdisciplinary clinical-research studies. In particular, the ABCB1 loci rs2032582 and rs1128503 showed significant associations in some studies ( Abraham et al., 2014 ; Lambrechts et al., 2015 ) but failed to replicate or demonstrated variable directionality in others ( Leskelä et al., 2011 ) ( Supplementary Tables S5, S6 ). ABCB1 encodes P-glycoprotein, a key ATP-dependent drug efflux transporter expressed in the intestine, liver, kidney, blood–nerve barrier, and peripheral neurons, where it may influence systemic taxane exposure, tissue drug accumulation, and neuronal susceptibility to toxicity. The inconsistent findings therefore likely reflect heterogeneity across study populations, treatment regimens, ancestry, and phenotyping methodologies.

Several other discovery variants reported across multiple research studies did not reach statistical significance or have not been independently replicated and are catalogued in Supplementary Table S7 .

Platinum-based agents

19 studies investigated the genetic associations with platinum-induced peripheral neuropathy (PIPN) including 1 GWAS ( Dolan et al., 2017 ) as well as multiple candidate gene studies (See Table 5 for study characteristics, for full list of studies see Supplementary Material 2 ). The GWAS undertaken by Dolan et al. (2017) identified rs7349683 within EPHA5 , a gene implicated in normal neuronal development as well as in the repair mechanisms activated after nerve injury ( Dolan et al., 2017 ). However, none of the SNPs reported in this GWAS were subsequently replicated in candidate gene analyses. This represents an important limitation of previous research, given the scarcity of GWAS specifically investigating PIPN. Most platinum-related studies have focused on other chemotherapy-related toxicities (e.g., neutropenia), which did not meet the exclusion criteria for this review.

Most of the studies that met the inclusion criteria enrolled individuals of European ancestry who were primarily diagnosed with colorectal cancer ( Table 5 ). There were some studies which included participants of East Asian ancestry although some of the studies did not specifically report any ancestry-based clinical information. The patient ages ranged from 18 years ( Dolan et al., 2017 ) to 88 years ( Cecchin et al., 2013 ), with a mean age of approximately 65 years across studies ( Table 5 ).

Supplementary File 1 summarizes SNPs identified in candidate gene association studies. Although candidate gene approaches are inherently more prone to bias than hypothesis-free GWAS, the variants that were identified through this method have been implicated across a number of important pathophysiological processes that are involved in the development of CIPN. However, further functional validation is needed as part of future translational studies.

The most extensively studied variant in platinum-induced peripheral neuropathy was GSTP1 rs1695 , which showed marked inconsistency across studies ( Supplementary File 1 ). GSTP1 encodes glutathione S-transferase Pi 1, a key phase II detoxification enzyme involved in conjugation of reactive electrophilic compounds with glutathione relevant to platinum derivates metabolism. Reported effect estimates varied substantially, ranging from protective (HR = 0.47, 95% CI 0.21–1.04, p = 0.03 ( Hong et al., 2011 )) to increased risk (OR = 6.08, 95% CI 1.15–32.18, p = 0.05 ( Kumamoto et al., 2013 )), with some studies reporting no significant association ( Chen et al., 2020 ; Hong et al., 2011 ; Katayanagi et al., 2019 ; Kumamoto et al., 2013 ; Lecomte et al., 2006 ). These discrepancies were observed across cohorts differing in cancer type, treatment regimen, and phenotype definition. Specifically, studies varied in the use of oxaliplatin-based combinations (e.g., FOLFOX), outcome measures (cumulative, severe, or chronic CIPN), grading instruments (different CTCAE versions, TNSc, or study-specific scales), and timing of assessment ( Chen et al., 2020 ; Hong et al., 2011 ; Katayanagi et al., 2019 ; Kumamoto et al., 2013 ; Lecomte et al., 2006 ). Together, these findings indicate that the direction and magnitude of the GSTP1 rs1695 association are not consistent across studies.

In a prospective candidate gene association undertaken by De Jong et al. (2023) , the authors found a significant association between TRPV1 rs879207 and the severity of PIPN ( De Jong et al., 2023 ). TRPV1 encodes a nociceptive ion channel involved in pain signalling, thermal sensitivity, and calcium-dependent neuronal excitability, making it a biologically plausible mediator of chemotherapy-induced neuropathic pain and sensory dysfunction. The G allele was linked to increased neuropathy severity (OR 5.2, 95% CI 2.1–12.8; p = 0.012) in a cohort of 320 patients. The TRPV1 protein encodes a non-selective cation channel that acts as the receptor for capsaicin and plays a key role in nociception and pain signaling ( Rosenbaum and Simon, 2007 ).

Multiple studies have implicated ATP-binding cassette (ABC) transporter family genes as contributing to the susceptibility of oxaliplatin-induced peripheral neuropathy ( Yang et al., 2021 ). Variants across ABCC1 , ABCC2 , and ABCC4 have been associated with differences in neuropathy severity, supporting a role for altered drug efflux and detoxification mechanisms in mediating neurotoxicity ( Supplementary File 1 ). Two polymorphisms in ABCC4 , rs1729786 and rs6492763 , were associated with opposing effects on neuropathy severity, with rs1729786 conferring a protective effect (OR 0.63, 95% CI 0.45–0.87; adjusted p = 0.006) and rs6492763 associated with increased risk (OR 1.42, 95% CI 1.02–1.98; adjusted p = 0.0406) ( Johnson et al., 2015 ). In ABCC1 , rs2074087 and rs35587 were both linked to reduced neuropathy severity, with effect estimates ranging from OR 0.43 (95% CI 0.22–0.86; p = 0.017) to OR 0.47 (95% CI 0.23–0.96; p = 0.0375) ( Cecchin et al., 2013 ).

Several ABCC2 polymorphisms were significantly associated with increased levels of neuropathy severity, including rs1885301 (OR 3.06, 95% CI 1.35–6.92; p = 0.0072), rs717620 (OR 14.39, 95% CI 1.63–127.02; p = 0.0164), rs4148396 (OR 4.69, 95% CI 1.60–13.74; p = 0.0048), and rs3740066 (OR 2.99, 95% CI 1.16–7.70; p = 0.0231). In contrast, rs2273697 was found to have a protective association (OR 0.44, 95% CI 0.20–0.98; p = 0.0043) ( Cecchin et al., 2013 ). Collectively, these findings highlight the importance of ABC transporter polymorphisms, particularly within ABCC2 , as potential modulators of platinum drug clearance as well as neuronal toxicity. The direction of effects across variants and the lack of replication across independent cohorts warrant further validation across larger, ethnically diverse populations.

Several single-study associations implicated genes involved in DNA repair pathways. The alkyl-adduct reversal enzyme MGMT exhibited a protective effect at rs11016884 (OR = 0.64, 95% CI 0.46–0.90, p = 0.0104) ( Johnson et al., 2015 ). Within the nucleotide excision repair pathway, ERCC1 rs11615 conferred increased neuropathy risk (OR = 2.4, 95% CI 1.1–5.4, p = 0.047) ( Chen et al., 2010 ). The homologous recombination mediator RAD51 carried a protective allele at rs3092981 (OR = 0.59, 95% CI 0.35–0.99, p = 0.0445) ( Johnson et al., 2015 ), whereas the mismatch repair component MSH3 rs26279 was associated with greater neuropathy risk (OR = 1.39, 95% CI 1.01–1.93, p = 0.0467) ( Johnson et al., 2015 ). These findings highlight the involvement of multiple DNA repair mechanisms, including nucleotide excision repair, homologous recombination, and mismatch repair, as biologically plausible pathways contributing to platinum-induced neurotoxicity ( Supplementary File 1 ).

Several other discovery variants reported across multiple studies did not reach statistical significance or have not been independently replicated and are catalogued in Supplementary Table S8 .

Bortezomib

Six studies were identified which specifically investigated the association between genetic polymorphisms and the risk of developing bortezomib-induced peripheral neuropathy (BIPN). The identified studies included 3 GWAS ( Campo et al., 2018 ; García-Sanz et al., 2017 ; Magrangeas et al., 2016 ), 1 GWAS replication cohort ( Campo et al., 2017 ), and two candidate gene studies ( Corthals et al., 2011 ; Zhang et al., 2024 ) (see Table 6 for study characteristics, for full list of studies see Supplementary Material 2 ). The relatively small number of bortezomib studies reflects a limited pharmacogenomic evidence base for this drug class rather than a search strategy constraint, as no restrictions on chemotherapeutic agents were applied ( Supplementary Material ). All studies were conducted in adult populations with multiple myeloma. Most of the cohorts included individuals predominantly of European ancestry although there was one study which only included individuals of Chinese ancestry ( Zhang et al., 2024 ). Two studies ( García-Sanz et al., 2017 ; Magrangeas et al., 2016 ) did not specify the ancestry of the participants but recruited individuals from the GEM05MAS65, HOVON-65, and IFM trials, which have been reported elsewhere and primarily include patients of European descent.

None of the SNPs identified in the GWAS setting achieved statistical significance upon replication in candidate gene studies ( Table 7 ). Nonetheless, several variants have been highlighted in candidate gene analyses, including MTHFR (rs1801131), which regulates folate metabolism and methylation pathways important for neuronal repair and oxidative stress responses; EDN1 (rs5370), encoding endothelin-1, a potent vasoactive peptide implicated in microvascular dysfunction and neuroinflammation; IL17RD (rs1545981), a modulator of interleukin-17 and inflammatory signalling pathways; and the ABC transporter gene ABCC6 (rs8058696), which may influence xenobiotic transport and tissue susceptibility to toxic metabolites. These loci demonstrated significant associations with BIPN across several independent studies. The odds ratios ranged from 1.58 ( EDN1 ) to 2.63 ( MTHFR ), with rs1801131 achieving p < 0.001 in a Chinese Han cohort ( Zhang et al., 2024 ), which was further supported by findings from a Dutch clinical cohort study ( Broyl et al., 2010 ). However, replication findings for EDN1 and IL17RD differed in effect direction, further highlighting the potential impact of population or phenotype heterogeneity ( Supplementary Table S9 ) ( Campo et al., 2017 ).

In terms of the GWAS findings, PKNOX1 rs2839629 was associated with an increased risk of grade ≥2 neuropathy (OR = 2.04, 95% CI 1.11–3.33; p = 0.036) ( Magrangeas et al., 2016 ). PKNOX1 encodes a transcription factor involved in gene regulation during neuronal development and inflammatory responses, supporting a plausible role in BIPN susceptibility. Additional loci included PRDM6 rs80146384, a regulator of vascular smooth muscle differentiation and vascular homeostasis ( García-Sanz et al., 2017 ), as well as TENM3-AS1/LINC02500 rs6552496 and DCC rs17748074, both linked to neuronal development and axon guidance pathways ( Campo et al., 2018 ). The latter GWAS also identified CDH13 rs8060632, encoding a cadherin involved in cell adhesion and neuronal connectivity, and CMYA5 rs12521798, a cytoskeletal-associated gene potentially relevant to structural resilience, surpassing suggestive thresholds, though none reached genome-wide significance or were validated in independent cohorts ( Table 7 ).

Several additional variants remain single-study findings, including MTHFR rs1801133 and rs17421511 ( Zhang et al., 2024 ), and ABCC1 rs2014800 ( Campo et al., 2018 ), where ABCC1 encodes a multidrug resistance transporter involved in cellular efflux of xenobiotics and oxidative metabolites. Previously proposed associations, such as intronic variants in RHOBTB2 , a regulator of cytoskeletal dynamics and vesicle trafficking, and PLCG2 rs45443101, involved in intracellular calcium and immune signalling, were not replicable upon reanalysis ( Broyl et al., 2010 ; García-Sanz et al., 2017 ).

Collectively, these results indicate biologically plausible roles for genes involved in axon guidance, vascular signalling, detoxification transport, folate metabolism, and inflammatory regulation, but underscore the lack of consistent replication across studies and the need for larger, multi-ancestry GWAS and functional validation efforts ( Tables 4 ; Supplementary Tables S9, S10 ).

Several recently discovered variants that were reported across multiple studies did not reach statistical significance and were not independently replicated and are catalogued in Supplementary Table S10 .

Additional studies included following manual search

Across taxanes, vinca alkaloids, and platinum-based therapies, 11 manually curated studies were also included in this review ( Adjei et al., 2021 ; Chua et al., 2020 ; Kanai et al., 2021 ; Kanai et al., 2016 ; Klumpers et al., 2022 ; Komatsu et al., 2015 ; Mahmoudpour et al., 2018 ; Martin-Guerrero et al., 2019 ; Min et al., 2024 ; Wright et al., 2019 ; Zgheib et al., 2018 ). For a comprehensive summary of significant and non-significant findings from these studies see Table 8 for study characteristics and Supplementary Tables S11, S12 .

Paclitaxel-associated CIPN showed the strongest and most numerous associations within this data additional set, including genome-wide significant loci and variants in genes involved in neuronal structure, cytoskeletal regulation, transcriptional control, and immune signaling ( Adjei et al., 2021 ; Chua et al., 2020 ; Mahmoudpour et al., 2018 ). Notably, rs74497159 in S1PR1 emerged as a robust paclitaxel-specific association, implicating sphingosine-1-phosphate signaling, an established regulator of endothelial function, neuroinflammation, and barrier integrity, as a mechanistically plausible and potentially druggable pathway in CIPN ( Chua et al., 2020 ).

For vincristine, fewer loci were reported, however similar to studies above, the most consistent evidence was observed for variants in CEP72 , a gene involved in chromosome segregation, across multiple cohorts and genetic models. Additional associations implicated genes related to purine metabolism, and inflammatory signaling, although findings were generally derived from smaller candidate gene studies ( Klumpers et al., 2022 ; Mahmoudpour et al., 2018 ; Wright et al., 2019 ).

In oxaliplatin-treated cohorts, GWAS identified multiple loci associated with neuropathy severity and recovery, mapping to genes involved in synaptic transmission, calcium signaling, metal transport, extracellular matrix integrity, and neuronal development ( Kanai et al., 2021 ). Several associations localized to non-coding regions, suggesting regulatory mechanisms contributing to inter-individual variability in both CIPN onset and resolution.

However, comparison of Supplementary Tables S11, S12 revealed additional substantial heterogeneity and inconsistency in these reported genetic associations with CIPN across studies and treatment regimens. Variants reported as significant in some cohorts frequently showed null associations, opposite directions of effect, or lack of independent replication in others.

For vincristine, CEP72 exemplified this discordance: while certain studies reported significant associations with neuropathy severity ( Supplementary Table S11 ), independent analyses of rs868649 and rs924607 showed no association ( Supplementary Table S12 ) ( Klumpers et al., 2022 ; Zgheib et al., 2018 ).

For paclitaxel, several loci highlighted in Supplementary Table S11 , including genes involved in cytoskeletal and neuronal processes (e.g., FGD4 ), demonstrated weaker or non-significant effects in curated null studies, and multiple previously proposed candidates in drug metabolism and transport pathways ( CYP2C8 , ABCC2 , ABCB1 ) failed to reach corrected significance thresholds ( Supplementary Table S12 ) ( Adjei et al., 2021 ; Komatsu et al., 2015 ).

In oxaliplatin-treated cohorts, variants in detoxification, DNA repair, and ion channel pathways (e.g., GSTP1 , ERCC1 , SCN10A ) were predominantly non-significant across multiple studies ( Supplementary Table S12 ) ( Adjei et al., 2021 ; Kanai et al., 2016 ). Similarly, for bortezomib, the majority of candidate variants spanning immune signaling, proteasome components, and transporter genes showed null associations ( Supplementary Table S12 ) ( Min et al., 2024 ).

Discussion

Chemotherapy-induced peripheral neuropathy (CIPN) remains a major dose-limiting toxicity of cancer treatment, with prevalence ranging from approximately 19% to over 85% depending on the chemotherapeutic agent ( Fallon, 2013 ; Starobova and Vetter, 2017 ). Despite its clinical significance and long-term impact on quality of life, there are currently no effective disease-modifying therapies ( Sałat, 2020 ). This systematic review synthesised evidence from 72 studies investigating germline genetic contributions to CIPN across vincristine, taxanes, platinum agents, and bortezomib. While numerous genetic associations have been reported, including replicated loci such as CEP72 and ETAA1 in vincristine-induced neuropathy ( Diouf et al., 2015 ; Van De Velde et al., 2022 ) and EPHA5 , FGD4 , and FZD3 in taxane-induced neuropathy ( Baldwin et al., 2012 ; Boora et al., 2015 ; Chen et al., 2020 ; Leandro-García et al., 2013 ; Tanabe et al., 2020 ) the overall evidence is characterised by limited reproducibility, inconsistent directionality of effects, and lack of clinical translation. Importantly, by integrating both positive and negative findings, this review moves beyond descriptive synthesis to critically evaluate the reliability of pharmacogenomic signals in CIPN and identifies the key barriers preventing translation into clinically actionable biomarkers.

Why pharmacogenomic findings in CIPN fail to replicate

Although numerous variants have been proposed, only a small subset has demonstrated consistent replication across independent cohorts, and none has yet achieved sufficient clinical validity for predictive implementation. Collectively, the available evidence supports the biological plausibility of several pathophysiological pathways involved in CIPN but also highlights substantial methodological limitations that hinder reproducibility and translation into clinical practice. Additionally, it is important to note that different chemotherapeutic agents act through distinct mechanisms, making genetic heterogeneity in CIPN risk factors likely.

The variability observed for GSTP1 rs1695 highlights a key limitation in CIPN pharmacogenomics. Despite being the most frequently studied variant in platinum-induced neuropathy, its effect direction is not stable across studies ( Chen et al., 2020 ; Hong et al., 2011 ; Katayanagi et al., 2019 ; Kumamoto et al., 2013 ; Lecomte et al., 2006 ). This inconsistency appears to be driven by substantial heterogeneity in study design ( Supplementary Table S13 ), including differences in cancer type, platinum regimen and cumulative exposure, phenotype definition (e.g., cumulative versus chronic neuropathy), assessment tools, and timing of evaluation ( Chen et al., 2020 ; Hong et al., 2011 ; Katayanagi et al., 2019 ; Kumamoto et al., 2013 ; Lecomte et al., 2006 ). Population differences may further contribute, given variability in allele frequencies and genetic background across cohorts. As a result, GSTP1 rs1695 is best interpreted as a context-dependent association, rather than a robust or generalisable biomarker. This finding underscores a broader issue in the field: biologically plausible pathways, such as detoxification and oxidative stress, are repeatedly implicated, but variant-level effects remain inconsistent due to unresolved heterogeneity in phenotyping and clinical context.

A major limitation to replication was substantial heterogeneity in study design, with CIPN assessed using diverse clinical scales and patient-reported measures, limiting comparability and increasing measurement error. Statistical methods varied widely, including different multiple-testing corrections or nominal p-values, and many studies inadequately adjusted for confounders, further reducing interpretability. Inconsistent allele annotation frequently complicated interpretation. In many reports, the effect allele was not explicitly specified, preventing harmonization and sometimes reversing inferred effect directions. As highlighted by Wootton and Sallis, (2020) , inconsistent terminology, such as “reference”, “alternate”, “minor”, or “A1/A2”, without clarification, undermines data integrity and limits reproducibility. Entries lacking clear annotation were coded as “Not reported” in this review, reflecting broader issues in pharmacogenomic reporting practices.

Another major limitation of the current evidence base is the strong overrepresentation of European ancestry populations, particularly in taxane studies where 19 of 25 cohorts were predominantly European ( Table 3 ). This is not a minor limitation but a critical barrier to generalisability. Allele frequencies for several commonly studied variants differ substantially across populations, and effect sizes estimated in European cohorts may not translate to non-European groups. For example, variants such as CYP3A5 , which is rare in European populations but common in African ancestry groups, and loci including EPHA5, FGD4 , and S1PR1 , identified primarily in European cohorts, may have different effect sizes or linkage structures in other populations. Similarly, CEP72 (vincristine) and GSTP1 rs1695 (platinum agents) show population-dependent allele frequencies that may influence both statistical power and observed direction of effect.

The limited representation of Asian, African, and other ancestries restricts the ability to determine whether observed associations are broadly applicable or population-specific. As a result, current pharmacogenomic findings cannot be assumed to be generalisable across global patient populations. This has direct implications for precision oncology, where applying European-derived genetic risk markers to diverse populations may lead to inaccurate risk prediction. Addressing this limitation will require multi-ancestry cohorts, replication in underrepresented populations, and explicit reporting of ancestry-stratified effects.

Cumulative chemotherapy exposure represents another source of heterogeneity across CIPN pharmacogenomic studies. As summarised in Supplementary Table S14 , adjustment for cumulative dose was inconsistently performed or reported across the included studies. This limitation is particularly relevant for taxane- and oxaliplatin-treated cohorts, where neuropathy severity and persistence are closely related to cumulative exposure. In unadjusted analyses, a variant that appears protective may instead identify patients who tolerated higher cumulative doses before developing neuropathy, whereas a risk variant may be confounded by treatment modification, early discontinuation, or differential exposure. Therefore, inconsistent dose adjustment likely contributes to non-replication and unstable effect directions across studies. Future pharmacogenomic studies should report planned and delivered cumulative dose, dose reductions, treatment delays, and discontinuation, and should include these exposure variables in sensitivity analyses where possible.

A further important and underappreciated source of heterogeneity across studies is the wide age range of included patient populations ( Tables 1 , 3 , 5 , 7 ). The studies reviewed span pediatric cohorts, particularly in vincristine-treated acute lymphoblastic leukemia, to older adult populations receiving taxanes or platinum-based therapies, with reported ages ranging from early childhood to over 85 years. These populations differ fundamentally in pharmacokinetics, drug metabolism, neuronal resilience, and comorbidity burden, all of which are known to influence CIPN risk and severity ( Starobova and Vetter, 2017 ; Tay et al., 2022 ). As a result, genetic associations identified in pediatric settings may not be directly comparable to those observed in adult or elderly populations. For example, variants influencing drug metabolism or neurodevelopmental pathways in children may have different functional consequences in the context of age-related neuronal vulnerability or cumulative physiological stress in older adults. Despite this, relatively few studies explicitly account for age as a modifier or perform age-stratified analyses. The absence of systematic consideration of age effects likely contributes to inconsistencies in the direction and magnitude of reported associations across studies. This highlights the need for future pharmacogenomic investigations to incorporate age as a key biological variable, including through stratified analyses and multi-cohort validation across distinct age groups.

Interpretation and reproducibility is also complicated by numerous variants of uncertain significance (VUS) requiring combined clinical, computational and functional evidence, with unclear contributions to CIPN risk. Integrative functional and population analyses are needed to establish pathogenicity, although many VUS will remain unresolved.

Additionally, small discovery cohorts with inflated effect estimates that fail replication point to a “winner’s curse,” and many reported variants lie in non-coding regions with unknown functional consequences, underscoring the need for integrative functional genomics.

Although publication bias is a recognized concern in CIPN pharmacogenomics, formal quantitative assessment (e.g., funnel plots or Egger’s regression) was not feasible because studies differed substantially in chemotherapy exposures, phenotype definitions, genetic models, and reporting metrics. Nevertheless, the literature demonstrates patterns consistent with selective reporting and winner’s curse, whereby numerous loci were significant only in discovery cohorts and showed attenuated, null, or inconsistent effects in later studies.

To address whether replication failure in CIPN pharmacogenomics is primarily methodological or biological, we incorporated Q-Genie scores into interpretation and stratified findings by study quality ( Table 9 ). Overall, 52.4% of studies were high quality and 37.7% moderate quality. Higher-quality studies were typically GWAS, prospective cohorts, or multicentre analyses, and identified the most credible replicated loci, including CEP72 and ETAA1 (vincristine), EPHA5 , FGD4 , FZD3 , SCN9A , and S1PR1 (taxanes), and recurrent GSTP1 and ABCC2 associations in platinum-treated cohorts. Importantly, Table 9 shows that non-replication was not limited to lower-quality studies. Several high-quality studies also reported null validation attempts, reduced effect sizes, or inconsistent directions of effect, particularly for ABCB1 , CYP2C8 , CYP3A5 , GSTP1 , and multiple GWAS loci. Thus, improved methodology alone does not fully explain reproducibility failures. Moderate-quality studies showed mixed positive and null findings, while lower-quality studies more often reported isolated candidate gene associations without independent replication, consistent with higher false-positive risk. Collectively, these findings suggest that replication failure reflects both methodological limitations and biological complexity, including heterogeneous CIPN phenotyping, ancestry differences, treatment variation, and modest true effect sizes. Future progress will require harmonised phenotyping, multi-ancestry cohorts, standardised reporting, and polygenic or pathway-level approaches rather than reliance on single SNP.

Biological convergence despite genetic heterogeneity

Despite inconsistent replication at the level of individual variants, a clear pattern of biological convergence emerges across studies. Genetic associations consistently implicate pathways related to neuronal structure and function, drug metabolism, and cellular stress responses.

Vincristine-induced peripheral neuropathy (VIPN), particularly relevant in pediatric acute lymphoblastic leukaemia (ALL), remains the most extensively studied phenotype. The most consistently replicated associations were found in CEP72 and ETAA1 , genes involved in microtubule formation and DNA damage-response pathways, respectively, both central to neuronal susceptibility ( Diouf et al., 2015 ; Van De Velde et al., 2022 ). Additional candidate genes such as NDRG1, GARS, and FGD4 , implicated in axonal growth and cytoskeletal maintenance, support a mechanistic model in which impaired neuronal structural integrity contributes to VIPN risk ( Van De Velde et al., 2022 ). MicroRNA polymorphisms (e.g., miR-3117, miR-4481 ) may act as upstream regulators of neuroplasticity-related targets, though require further validation ( Gutierrez-Camino et al., 2018 ). Within the manually curated dataset, fewer loci were reported for vincristine, with the most consistent evidence again observed for variants in CEP72 ( Klumpers et al., 2022 ; Mahmoudpour et al., 2018 ; Wright et al., 2019 ). However, direct comparison revealed substantial heterogeneity, with rs924607 and rs868649 showing null or inconsistent associations after covariate adjustment in curated analyses.

For taxane-induced peripheral neuropathy (TIPN), replicated associations were observed in EPHA5, EPHA6, and FZD3 , all involved in axon guidance and neurite development, contrasting with variants in SCN9A , which relate to altered ion channel excitability ( Baldwin et al., 2012 ; Boora et al., 2015 ; Chen et al., 2020 ; Leandro-García et al., 2013 ; Tanabe et al., 2020 ). The repeated identification of FGD4 across both vincristine and taxane cohorts suggests overlapping microtubule-dependent mechanisms of axonal maintenance ( Baldwin et al., 2012 ; Van De Velde et al., 2022 ). Variants in pharmacokinetic genes such as CYP2C8, CYP3A5 and ABCB1 influence drug metabolism and efflux, although their effects differ across populations and treatment protocols, emphasizing the need to integrate pharmacokinetic and neuronal susceptibility factors in predictive models ( Abraham et al., 2014 ; Boora et al., 2015 ; Leskelä et al., 2011 ; Rizzo et al., 2010 ; Vargas-Aliaga et al., 2024 ). Within the manually curated set, TIPN showed the strongest and most numerous genetic associations, including genome-wide significant loci and variants mapping to genes involved in neuronal structure, cytoskeletal regulation, transcriptional control, and immune signalling ( Adjei et al., 2021 ; Chua et al., 2020 ; Mahmoudpour et al., 2018 ). Notably, the association of rs74497159 in S1PR1 highlights sphingosine-1-phosphate signalling as a biologically plausible and potentially druggable pathway in paclitaxel-induced CIPN ( Chua et al., 2020 ). However, replicated signals in EPHA5, FZD3 , and FGD4 were not uniformly supported, with several variants showing weaker effects, loss of significance after correction, or variable effect direction. In contrast, S1PR1 was largely absent from earlier studies identified through database and register search.

Evidence for platinum-induced peripheral neuropathy (PIPN) is derived predominantly from candidate gene studies. Recurrent associations with ABCC1, ABCC2, ABCC4 , and GSTP1 highlight impaired detoxification and reduced efflux of platinum compounds in the context of neuronal oxidative stress ( Cecchin et al., 2013 ; Chen et al., 2020 ; Hong et al., 2011 ; Johnson et al., 2015 ; Katayanagi et al., 2019 ; Kumamoto et al., 2013 ; Lecomte et al., 2006 ). Variants in DNA repair genes, including ERCC1, RAD51, MSH3, and MGMT, modulate risk, consistent with the neurotoxic effects of unrepaired DNA crosslinks ( Chen et al., 2010 ; Johnson et al., 2015 ). The recently identified TRPV1 rs879207 variant suggests a mechanistic link between nociceptive signalling and platinum-induced hypersensitivity, though functional studies are needed ( De Jong et al., 2023 ). Within the curated dataset, oxaliplatin studies identified multiple loci linked to neuropathy severity and recovery, implicating synaptic, calcium, and extracellular matrix–related pathways ( Klumpers et al., 2022 ; Mahmoudpour et al., 2018 ; Wright et al., 2019 ). However, variants in GSTP1, ABCC2 , and DNA repair genes ( ERCC1, XRCC1 ) demonstrated marked variability across both data sets, with effect directions ranging from protective to deleterious.

For bortezomib-induced peripheral neuropathy (BIPN), evidence remains comparatively limited. Although several GWAS have reported suggestive loci (e.g., PKNOX1, PRDM6, DCC ), none has achieved genome-wide significance or been independently validated ( Campo et al., 2018 ; García-Sanz et al., 2017 ; Magrangeas et al., 2016 ). Candidate gene studies more consistently implicate MTHFR (rs1801131), EDN1 (rs5370), IL17RD (rs1545981) and ABCC6 (rs8058696), pointing to roles in folate metabolism, vascular regulation, inflammatory signalling and detoxification ( Broyl et al., 2010 ; Campo et al., 2017 ; Zhang et al., 2024 ). Inconsistent effect directions may reflect ancestry-related allele frequency differences and phenotype heterogeneity. Notably, the repeated involvement of ABCC family genes across platinum and bortezomib studies suggests a shared detoxification-related pathway contributing to neurotoxicity ( Campo et al., 2018 ; Zhang et al., 2024 ).

Across all drug classes, the most replicated loci, including CEP72, ETAA1, EPHA5, FGD4 and ABCC2 , converge on pathways related to microtubule integrity, axon guidance, DNA repair and drug efflux. Together, these genes support a polygenic model of CIPN characterized by axonal degeneration driven by oxidative stress, disrupted calcium homeostasis and impaired cytoskeletal stability ( Starobova and Vetter, 2017 ).

Importantly, while individual variants often show inconsistent replication or opposing effects across studies, the recurrence of these biological pathways suggests that CIPN susceptibility reflects a polygenic architecture with pathway-level convergence. This indicates that focusing on individual SNPs may be insufficient to capture the complexity of CIPN risk, and that pathway-based or systems-level approaches may provide greater insight. Although polygenic risk scores (PRS) have been explored in recent work ( Hooshmand et al., 2022 ), they remain uncommon. PRS approaches may eventually provide valuable insights into cumulative genetic risk across drug classes, provided they are developed using harmonized, multi-ancestry datasets.

Implications for pharmacogenomic biomarker development

A key implication of these findings is that no genetic variants currently meet criteria for clinical implementation in CIPN risk prediction. Despite extensive investigation, the lack of robust, reproducible associations limits the utility of pharmacogenomic testing in guiding treatment decisions or identifying high-risk patients.

To overcome these limitations, future studies must adopt more rigorous and standardised approaches. Harmonisation of phenotyping is essential, including consistent use of validated neuropathy assessment tools and clear definitions of clinical endpoints. Large, well-powered multi-ancestry GWAS are required to improve reproducibility and identify variants with consistent effects across populations.

In addition, standardised reporting practices, including clear annotation of effect alleles and statistical thresholds, are critical to enable cross-study comparisons and meta-analysis. Functional validation of candidate variants is also needed to establish biological relevance and distinguish true causal relationships from statistical associations.

Finally, integration of genetic data with other molecular and clinical datasets through multi-omics approaches offers a promising strategy to improve predictive accuracy. Combining genomic, transcriptomic, proteomic, and clinical data may enable the development of more robust models for risk stratification and precision oncology in CIPN.

Importantly, even if all reported risk alleles are true, their small effect sizes explain only a limited proportion of CIPN risk, with known loci likely accounting for no more than ∼5–15% of variance, making genetics alone insufficient for clinical prediction. This level of variance is inadequate for high-stakes decisions such as dose reduction but may be useful for lower-risk applications like trial enrichment or targeted monitoring. Overall, these findings support a highly polygenic, multifactorial CIPN architecture requiring integrated models combining genetics with clinical, pharmacokinetic and biological factors ( Hooshmand et al., 2022 ; Starobova and Vetter, 2017 ).

Conclusions

In summary, this systematic review demonstrates that while numerous genetic associations with CIPN have been reported, the field is characterised by poor replication, methodological heterogeneity, and limited translational progress. However, the convergence of findings across key biological pathways supports a polygenic and mechanistically coherent model of CIPN susceptibility. Advancing pharmacogenomics in this field will require coordinated efforts to standardise study design, expand cohort diversity, and integrate functional and multi-omics data to enable clinically actionable precision medicine approaches.

Acknowledgements

We thank Ms. Andree Axelsen for financial support of this study.

Funding Statement

The author(s) declared that financial support was received for this work and/or its publication. HS was supported by NHMRC Investigator grant 2024/GNT2034754, IV was supported by NHMRC Investigator grant 2017086.

Footnotes

Edited by: R. Stephanie Huang , University of Minnesota Twin Cities, United States

Reviewed by: Yuting Shan , University of Minnesota, United States

Suneeta Modekurty , SANJEEVANI AI LLC, United States

Lydia Hach , University of Minnesota Twin Cities, United States

Data availability statement

The original contributions presented in the study are included in the article/ Supplementary Material , further inquiries can be directed to the corresponding authors.

Author contributions

AM: Conceptualization, Data curation, Formal Analysis, Methodology, Validation, Visualization, Writing – original draft. IR: Data curation, Formal Analysis, Writing – review and editing. HuS: Data curation, Formal Analysis, Writing – review and editing. EK: Data curation, Formal Analysis, Writing – review and editing. JB: Data curation, Formal Analysis, Writing – review and editing. NR: Data curation, Formal Analysis, Writing – review and editing. SS: Data curation, Formal Analysis, Writing – review and editing. KS: Data curation, Formal Analysis, Writing – review and editing. PZ: Data curation, Formal Analysis, Writing – review and editing. SP: Conceptualization, Methodology, Supervision, Validation, Visualization, Writing – original draft, Writing – review and editing. DS: Conceptualization, Data curation, Formal Analysis, Methodology, Supervision, Writing – review and editing. E-LL: Conceptualization, Data curation, Formal Analysis, Methodology, Supervision, Writing – review and editing. LY: Conceptualization, Data curation, Methodology, Supervision, Validation, Writing – review and editing. IV: Conceptualization, Funding acquisition, Methodology, Resources, Supervision, Visualization, Writing – original draft, Writing – review and editing. HS: Conceptualization, Data curation, Formal Analysis, Funding acquisition, Investigation, Methodology, Resources, Supervision, Validation, Visualization, Writing – original draft, Writing – review and editing.

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Supplementary material

The Supplementary Material for this article can be found online at: https://www.frontiersin.org/articles/10.3389/fphar.2026.1842379/full#supplementary-material

References

Abdelfattah N. M., Solayman M. H., Elnahass Y., Sabri N. A. (2021). ABCB1 single nucleotide polymorphism genotypes as predictors of paclitaxel-induced peripheral neuropathy in breast cancer. Genet. Test. Mol. Biomarkers 25 (7), 471–477. 10.1089/gtmb.2021.0014

Abraham J. E., Guo Q., Dorling L., Tyrer J., Ingle S., Hardy R., et al. (2014). Replication of genetic polymorphisms reported to be associated with taxane-related sensory neuropathy in patients with early breast cancer treated with paclitaxel. Clin. Cancer Res. 20 (9), 2466–2475. 10.1158/1078-0432.CCR-13-3232

Adjei A. A., Lopez C. L., Schaid D. J., Sloan J. A., Le-Rademacher J. G., Loprinzi C. L., et al. (2021). Genetic predictors of chemotherapy-induced peripheral neuropathy from paclitaxel, carboplatin and oxaliplatin: ncctg/alliance N08C1, N08CA and N08CB study. Cancers 13 (5), 1084. 10.3390/cancers13051084

Antonacopoulou A. G., Argyriou A. A., Scopa C. D., Kottorou A., Kominea A., Peroukides S., et al. (2010). Integrin beta‐3 L33P: a new insight into the pathogenesis of chronic oxaliplatin‐induced peripheral neuropathy? Euro J Neurology 17 (7), 963–968. 10.1111/j.1468-1331.2010.02966.x

Aplenc R., Glatfelter W., Han P., Rappaport E., La M., Cnaan A., et al. (2003). CYP3A genotypes and treatment response in paediatric acute lymphoblastic leukaemia. Br. J. Haematol. 122 (2), 240–244. 10.1046/j.1365-2141.2003.04430.x

Arbitrio M., Scionti F., Di Martino M. T., Caracciolo D., Pensabene L., Tassone P., et al. (2021). Pharmacogenomics biomarker discovery and validation for translation in clinical practice. Clin. Transl. Sci. 14 (1), 113–119. 10.1111/cts.12869

Areti A., Yerra V. G., Naidu V., Kumar A. (2014). Oxidative stress and nerve damage: role in chemotherapy induced peripheral neuropathy. Redox Biol. 2, 289–295. 10.1016/j.redox.2014.01.006

Argyriou A. A., Polychronopoulos P., Iconomou G., Koutras A., Kalofonos H. P. (2008). A review on oxaliplatin-induced peripheral nerve damage. Cancer Treat. Rev. 34 (4), 368–377. 10.1016/j.ctrv.2008.01.003

Baldwin R. M., Owzar K., Zembutsu H., Chhibber A., Kubo M., Jiang C., et al. (2012). A genome-wide association study identifies novel loci for paclitaxel-induced sensory peripheral neuropathy in CALGB 40101. Clin. Cancer Res. 18 (18), 5099–5109. 10.1158/1078-0432.CCR-12-1590

Basso M., Modoni A., Spada D., Cassano A., Schinzari G., Lo Monaco M., et al. (2011). Polymorphism of CAG motif of SK3 gene is associated with acute oxaliplatin neurotoxicity. Cancer Chemother. Pharmacol. 67 (5), 1179–1187. 10.1007/s00280-010-1466-y

Bergmann T. K., Brasch‐Andersen C., Gréen H., Mirza M. R., Skougaard K., Wihl J., et al. (2012). Impact of ABCB1 variants on neutrophil depression: a pharmacogenomic study of paclitaxel in 92 women with ovarian cancer. Basic Clin. Pharma Tox 110 (2), 199–204. 10.1111/j.1742-7843.2011.00802.x

Bergmann T. K., Vach W., Feddersen S., Eckhoff L., Gréen H., Herrstedt J., et al. (2013). GWAS-based association between RWDD3 and TECTA variants and paclitaxel induced neuropathy could not be confirmed in Scandinavian ovarian cancer patients. Acta Oncol. 52 (4), 871–873. 10.3109/0284186X.2012.707787

Boora G. K., Kulkarni A. A., Kanwar R., Beyerlein P., Qin R., Banck M. S., et al. (2015). Association of the Charcot–Marie–Tooth disease gene ARHGEF10 with paclitaxel induced peripheral neuropathy in NCCTG N08CA (Alliance). J. Neurological Sci. 357 (1-2), 35–40. 10.1016/j.jns.2015.06.056

Bosó V., Herrero M. J., Santaballa A., Palomar L., Megias J. E., De La Cueva H., et al. (2014). Snps and taxane toxicity in breast cancer patients. Pharmacogenomics 15 (15), 1845–1858. 10.2217/pgs.14.127

Brown M. R., Ramirez J. D., Farquhar-Smith P. (2014). Pain in cancer survivors. Br. J. Pain 8 (4), 139–153. 10.1177/2049463714542605

Broyl A., Corthals S. L., Jongen J. L., Van Der Holt B., Kuiper R., De Knegt Y., et al. (2010). Mechanisms of peripheral neuropathy associated with bortezomib and vincristine in patients with newly diagnosed multiple myeloma: a prospective analysis of data from the HOVON-65/GMMG-HD4 trial. Lancet Oncol. 11 (11), 1057–1065. 10.1016/S1470-2045(10)70206-0

Campo C., Da Silva Filho M. I., Weinhold N., Goldschmidt H., Hemminki K., Merz M., et al. (2017). Genetic susceptibility to Bortezomib-Induced peripheral neuroropathy: replication of the reported candidate susceptibility loci. Neurochem. Res. 42 (3), 925–931. 10.1007/s11064-016-2007-9

Campo C., Da Silva Filho M. I., Weinhold N., Mahmoudpour S. H., Goldschmidt H., Hemminki K., et al. (2018). Bortezomib‐induced peripheral neuropathy: a genome‐wide association study on multiple myeloma patients. Hematol. Oncol. 36 (1), 232–237. 10.1002/hon.2391

Cavaletti G., Jakubowiak A. J. (2010). Peripheral neuropathy during bortezomib treatment of multiple myeloma: a review of recent studies. Leukemia and Lymphoma 51 (7), 1178–1187. 10.3109/10428194.2010.483303

Cecchin E., D'Andrea M., Lonardi S., Zanusso C., Pella N., Errante D., et al. (2013). A prospective validation pharmacogenomic study in the adjuvant setting of colorectal cancer patients treated with the 5-fluorouracil/leucovorin/oxaliplatin (FOLFOX4) regimen. Pharmacogenomics J. 13 (5), 403–409. 10.1038/tpj.2012.31

Chan A., Hertz D. L., Morales M., Adams E. J., Gordon S., Tan C. J., et al. (2019). Biological predictors of chemotherapy-induced peripheral neuropathy (CIPN): MASCC neurological complications working group overview. Support. Care Cancer 27 (10), 3729–3737. 10.1007/s00520-019-04987-8

Chelala C., Khan A., Lemoine N. R. (2009). SNPnexus: a web database for functional annotation of newly discovered and public domain single nucleotide polymorphisms. Bioinformatics 25 (5), 655–661. 10.1093/bioinformatics/btn653

Chen Y. C., Tzeng C. H., Chen P. M., Lin J. K., Lin T. C., Chen W. S., et al. (2010). Influence of GSTP1 I105V polymorphism on cumulative neuropathy and outcome of FOLFOX‐4 treatment in Asian patients with colorectal carcinoma. Cancer Science 101 (2), 530–535. 10.1111/j.1349-7006.2009.01418.x

Chen Y., Fang F., Kidwell K. M., Vangipuram K., Marcath L. A., Gersch C. L., et al. (2020). Genetic variation in charcot–marie–tooth genes contributes to sensitivity to paclitaxel-induced peripheral neuropathy. Pharmacogenomics 21 (12), 841–851. 10.2217/pgs-2020-0053

Cho H.-J., Eom H.-S., Kim H.-J., Kim I.-S., Lee G. W., Kong S.-Y. (2010). Glutathione-S-transferase genotypes influence the risk of chemotherapy-related toxicities and prognosis in Korean patients with diffuse large B-cell lymphoma. Cancer Genet. Cytogenet. 198 (1), 40–46. 10.1016/j.cancergencyto.2009.12.004

Christofyllakis K., Kaddu-Mulindwa D., Lesan V., Rixecker T., Kos I. A., Held G., et al. (2024). An inherited genetic variant of the CEP72 gene is associated with the development of vincristine-induced peripheral neuropathy in female patients with aggressive B-cell lymphoma. Ann. Hematol. 103 (11), 4599–4606. 10.1007/s00277-024-05973-9

Chua W., Goldstein D., Lee C. K., Dhillon H., Michael M., Mitchell P., et al. (2009). Molecular markers of response and toxicity to FOLFOX chemotherapy in metastatic colorectal cancer. Br. J. Cancer 101 (6), 998–1004. 10.1038/sj.bjc.6605239

Chua K. C., Xiong C., Ho C., Mushiroda T., Jiang C., Mulkey F., et al. (2020). Genomewide meta‐analysis validates a role for S1PR1 in microtubule targeting agent‐induced sensory peripheral neuropathy. Clin Pharma Ther. 108 (3), 625–634. 10.1002/cpt.1958

Clark J. M., Sanders S., Carter M., Honeyman D., Cleo G., Auld Y., et al. (2020). Improving the translation of search strategies using the polyglot search translator: a randomized controlled trial. Jmla 108 (2), 195–207. 10.5195/jmla.2020.834

Cliff J., Jorgensen A. L., Lord R., Azam F., Cossar L., Carr D. F., et al. (2017). The molecular genetics of chemotherapy–induced peripheral neuropathy: a systematic review and meta-analysis. Crit. Rev. Oncology/Hematology 120, 127–140. 10.1016/j.critrevonc.2017.09.009

Corthals S. L., Kuiper R., Johnson D. C., Sonneveld P., Hajek R., Van Der Holt B., et al. (2011). Genetic factors underlying the risk of bortezomib induced peripheral neuropathy in multiple myeloma patients. Haematologica 96 (11), 1728–1732. 10.3324/haematol.2011.041434

Custodio A., Moreno-Rubio J., Aparicio J., Gallego-Plazas J., Yaya R., Maurel J., et al. (2014). Pharmacogenetic predictors of severe peripheral neuropathy in colon cancer patients treated with oxaliplatin-based adjuvant chemotherapy: a GEMCAD group study. Ann. Oncol. 25 (2), 398–403. 10.1093/annonc/mdt546

Dayem Ullah A. Z., Lemoine N. R., Chelala C. (2012). SNPnexus: a web server for functional annotation of novel and publicly known genetic variants (2012 update). Nucleic Acids Res. 40 (W1), W65–W70. 10.1093/nar/gks364

Dayem Ullah A. Z., Lemoine N. R., Chelala C. (2013). A practical guide for the functional annotation of genetic variations using SNPnexus. Briefings Bioinforma. 14 (4), 437–447. 10.1093/bib/bbt004

Dayem Ullah A. Z., Oscanoa J., Wang J., Nagano A., Lemoine N. R., Chelala C. (2018). SNPnexus: assessing the functional relevance of genetic variation to facilitate the promise of precision medicine. Nucleic Acids Res. 46 (W1), W109–W113. 10.1093/nar/gky399

De Graan A.-J. M., Elens L., Sprowl J. A., Sparreboom A., Friberg L. E., Van Der Holt B., et al. (2013). CYP3A4*22 genotype and systemic exposure affect paclitaxel-induced neurotoxicity. Clin. Cancer Res. 19 (12), 3316–3324. 10.1158/1078-0432.CCR-12-3786

De Jong C., Herder G. J. M., Van Haarlem S. W. A., Van Der Meer F. S., Van Lindert A. S. R., Ten Heuvel A., et al. (2023). Association between genetic variants and peripheral neuropathy in patients with NSCLC treated with first-line platinum-based therapy. Genes 14 (1), 170. 10.3390/genes14010170

Deng F., Laasik M., Salminen L., Lapatto L., Huhtinen K., Li Y., et al. (2023). Toxicity and therapy outcome associations in LIG3 , SLCO1B3 , ABCB1 , OPRM1 and GSTP1 in high‐grade serous ovarian cancer. Basic Clin. Pharma Tox 132 (6), 517–527. 10.1111/bcpt.13866

Diouf B., Crews K. R., Lew G., Pei D., Cheng C., Bao J., et al. (2015). Association of an inherited genetic variant with vincristine-related peripheral neuropathy in children with acute lymphoblastic leukemia. JAMA 313 (8), 815–823. 10.1001/jama.2015.0894

Dolan M. E., El C. O., Wheeler H. E., Gamazon E. R., Ardeshir-Rouhani-Fard S., Monahan P., et al. (2017). Clinical and genome-wide analysis of cisplatin-induced peripheral neuropathy in survivors of adult-onset cancer. Clin. Cancer Res. 23 (19), 5757–5768. 10.1158/1078-0432.ccr-16-3224

Egbelakin A., Ferguson M. J., MacGill E. A., Lehmann A. S., Topletz A. R., Quinney S. K., et al. (2011). Increased risk of vincristine neurotoxicity associated with low CYP3A5 expression genotype in children with acute lymphoblastic leukemia. Pediatr. Blood and Cancer 56 (3), 361–367. 10.1002/pbc.22845

Fallon M. (2013). Neuropathic pain in cancer. Br. Journal Anaesthesia 111 (1), 105–111. 10.1093/bja/aet208

García-Sanz R., Corchete L. A., Alcoceba M., Chillon M. C., Jiménez C., Prieto I., et al. (2017). Prediction of peripheral neuropathy in multiple myeloma patients receiving bortezomib and thalidomide: a genetic study based on a single nucleotide polymorphism array. Hematol. Oncol. 35 (4), 746–751. 10.1002/hon.2337

Guilhaumou R., Solas C., Bourgarel-Rey V., Quaranta S., Rome A., Simon N., et al. (2011). Impact of plasma and intracellular exposure and CYP3A4, CYP3A5, and ABCB1 genetic polymorphisms on vincristine-induced neurotoxicity. Cancer Chemother. Pharmacol. 68 (6), 1633–1638. 10.1007/s00280-011-1745-2

Glare P. A., Davies P. S., Finlay E., Gulati A., Lemanne D., Moryl N., et al. (2014). Pain in cancer survivor. J. Clinical Oncology 32 (16), 1739–1747. 10.1200/JCO.2013.52.4629

Goodenough C. G., Diouf B., Yang W., Sapkota Y., Finch E. R., Lu L., et al. (2022). Association between CEP72 genotype and persistent neuropathy in survivors of childhood acute lymphoblastic leukemia. Leukemia 36 (4), 1160–1163. 10.1038/s41375-021-01484-y

Gutierrez-Camino Á., Umerez M., Martin-Guerrero I., García De Andoin N., Santos B., Sastre A., et al. (2018). Mir-pharmacogenetics of vincristine and peripheral neurotoxicity in childhood B-cell acute lymphoblastic leukemia. Pharmacogenomics J. 18 (6), 704–712. 10.1038/s41397-017-0003-3

Hekmatnia Y., Movahednia N., Hajhamidiasl N., Hekmat E., Hekmat A., Khademi S. (2022). A review of the neurological complications of breast cancer. J. Fam. Med. Prim. Care 11 (8), 4205–4214. 10.4103/jfmpc.jfmpc_580_21

Henningsson A., Marsh S., Loos W. J., Karlsson M. O., Garsa A., Mross K., et al. (2005). Association of CYP2C8, CYP3A4, CYP3A5, and ABCB1 polymorphisms with the pharmacokinetics of paclitaxel. Clin. Cancer Research 11 (22), 8097–8104. 10.1158/1078-0432.CCR-05-1152

Hertz D. L., Motsinger-Reif A. A., Drobish A., Winham S. J., McLeod H. L., Carey L. A., et al. (2012). CYP2C8*3 predicts benefit/risk profile in breast cancer patients receiving neoadjuvant paclitaxel. Breast Cancer Res. Treat. 134 (1), 401–410. 10.1007/s10549-012-2054-0

Hertz D. L., Roy S., Motsinger-Reif A. A., Drobish A., Clark L. S., McLeod H. L., et al. (2013). CYP2C8*3 increases risk of neuropathy in breast cancer patients treated with paclitaxel. Ann. Oncol. 24 (6), 1472–1478. 10.1093/annonc/mdt018

Hong J., Han S.-W., Ham H. S., Kim T.-Y., Choi I. S., Kim B.-S., et al. (2011). Phase II study of biweekly S-1 and oxaliplatin combination chemotherapy in metastatic colorectal cancer and pharmacogenetic analysis. Cancer Chemother. Pharmacol. 67 (6), 1323–1331. 10.1007/s00280-010-1425-7

Hooshmand K., Goldstein D., Timmins H. C., Li T., Harrison M., Friedlander M. L., et al. (2022). Polygenic risk of paclitaxel-induced peripheral neuropathy: a genome-wide association study. J. Transl. Med. 20 (1), 564. 10.1186/s12967-022-03754-4

Johnson C., Pankratz V. S., Velazquez A. I., Aakre J. A., Loprinzi C. L., Staff N. P., et al. (2015). Candidate pathway-based genetic association study of platinum and platinum–taxane related toxicity in a cohort of primary lung cancer patients. J. Neurological Sci. 349 (1-2), 124–128. 10.1016/j.jns.2014.12.041

Jones D., Zhao F., Brell J., Lewis M. A., Loprinzi C. L., Weiss M., et al. (2015). Neuropathic symptoms, quality of life, and clinician perception of patient care in medical oncology outpatients with colorectal, breast, lung, and prostate cancer. J. Cancer Surviv 9 (1), 1–10. 10.1007/s11764-014-0379-x

Kanai M., Yoshioka A., Tanaka S., Nagayama S., Matsumoto S., Nishimura T., et al. (2010). Associations between glutathione S-transferase π Ile105Val and glyoxylate aminotransferase Pro11Leu and Ile340Met polymorphisms and early-onset oxaliplatin-induced neuropathy. Cancer Epidemiol. 34 (2), 189–193. 10.1016/j.canep.2010.02.008

Kanai M., Kawaguchi T., Kotaka M., Shinozaki K., Touyama T., Manaka D., et al. (2016). Large-scale prospective pharmacogenomics study of oxaliplatin-induced neuropathy in colon cancer patients enrolled in the JFMC41-1001-C2 (JOIN trial). Ann. Oncol. 27 (6), 1143–1148. 10.1093/annonc/mdw074

Kanai M., Kawaguchi T., Kotaka M., Manaka D., Hasegawa J., Takagane A., et al. (2021). Large-scale prospective genome-wide association study of oxaliplatin in stage II/III colon cancer and neuropathy. Ann. Oncol. 32 (11), 1434–1441. 10.1016/j.annonc.2021.08.1745

Katayanagi S., Katsumata K., Mori Y., Narahara K., Shigoka M., Matsudo T., et al. (2019). GSTP1 as a potential predictive factor for adverse events associated with platinum-based antitumor agent-induced peripheral neuropathy. Oncol. Lett. 17 (3), 2897–2904. 10.3892/ol.2019.9907

Kishi S., Cheng C., French D., Pei D., Das S., Cook E. H., et al. (2007). Ancestry and pharmacogenetics of antileukemic drug toxicity. Blood 109 (10), 4151–4157. 10.1182/blood-2006-10-054528

Klumpers M. J., Brand A. C., Hakobjan M., Gattuso G., Schiavello E., Terenziani M., et al. (2022). Contribution of common and rare genetic variants in CEP72 on vincristine‐induced peripheral neuropathy in brain tumour patients. Br. J. Clin. Pharmacol. 88 (7), 3463–3473. 10.1111/bcp.15267

Komatsu M., Wheeler H. E., Chung S., Low S.-K., Wing C., Delaney S. M., et al. (2015). Pharmacoethnicity in paclitaxel-induced sensory peripheral neuropathy. Clin. Cancer Res. 21 (19), 4337–4346. 10.1158/1078-0432.CCR-15-0133

Kulkarni A. A., Boora G., Kanwar R., Ruddy K. J., Banck M. S., Le-Lindqwister N., et al. (2015). RWDD3 and TECTA variants not linked to paclitaxel induced peripheral neuropathy in North American trial Alliance N08C1. Acta Oncol. 54 (8), 1227–1229. 10.3109/0284186X.2014.985388

Kumamoto K., Ishibashi K., Okada N., Tajima Y., Kuwabara K., Kumagai Y., et al. (2013). Polymorphisms of GSTP1, ERCC2 and TS-3′UTR are associated with the clinical outcome of mFOLFOX6 in colorectal cancer patients. Oncol. Lett. 6 (3), 648–654. 10.3892/ol.2013.1467

Kus T., Aktas G., Kalender M. E., Demiryurek A. T., Ulasli M., Oztuzcu S., et al. (2016). Polymorphism of CYP3A4 and ABCB1 genes increase the risk of neuropathy in breast cancer patients treated with paclitaxel and docetaxel. OncoTargets Therapy 9, 5073–5080. 10.2147/OTT.S106574

Kweekel D. M., Gelderblom H., Antonini N. F., Van Der Straaten T., Nortier J. W. R., Punt C. J. A., et al. (2009). Glutathione-S-transferase pi (GSTP1) codon 105 polymorphism is not associated with oxaliplatin efficacy or toxicity in advanced colorectal cancer patients. Eur. J. Cancer 45 (4), 572–578. 10.1016/j.ejca.2008.10.015

Lai J. I., Tzeng C. H., Chen P. M., Lin J. K., Lin T. C., Chen W. S., et al. (2009). Very low prevalence of XPD K751Q polymorphism and its association with XPD expression and outcomes of FOLFOX‐4 treatment in Asian patients with colorectal carcinoma. Cancer Sci. 100 (7), 1261–1266. 10.1111/j.1349-7006.2009.01186.x

Lambrechts S., Lambrechts D., Despierre E., Van Nieuwenhuysen E., Smeets D., Debruyne P. R., et al. (2015). Genetic variability in drug transport, metabolism or DNA repair affecting toxicity of chemotherapy in ovarian cancer. BMC Pharmacol. Toxicol. 16 (1), 2. 10.1186/s40360-015-0001-5

Leandro-García L. J., Inglada-Pérez L., Pita G., Hjerpe E., Leskelä S., Jara C., et al. (2013). Genome-wide association study identifies ephrin type A receptors implicated in paclitaxel induced peripheral sensory neuropathy. J. Med. Genet. 50 (9), 599–605. 10.1136/jmedgenet-2012-101466

Lecomte T., Landi B., Beaune P., Laurent-Puig P., Loriot M.-A. (2006). Glutathione S -Transferase P1 polymorphism (Ile105Val) predicts cumulative neuropathy in patients receiving oxaliplatin-based chemotherapy. Clin. Cancer Res. 12 (10), 3050–3056. 10.1158/1078-0432.CCR-05-2076

Leskelä S., Jara C., Leandro-García L. J., Martínez A., García-Donas J., Hernando S., et al. (2011). Polymorphisms in cytochromes P450 2C8 and 3A5 are associated with paclitaxel neurotoxicity. Pharmacogenomics J. 11 (2), 121–129. 10.1038/tpj.2010.13

Magrangeas F., Kuiper R., Avet-Loiseau H., Gouraud W., Guérin-Charbonnel C., Ferrer L., et al. (2016). A genome-wide association study identifies a novel locus for Bortezomib-Induced peripheral neuropathy in European patients with multiple myeloma. Clin. Cancer Res. 22 (17), 4350–4355. 10.1158/1078-0432.CCR-15-3163

Mahmoudpour S. H., Bandapalli O. R., da Silva Filho M. I., Campo C., Hemminki K., Goldschmidt H., et al. (2018). Chemotherapy-induced peripheral neuropathy: evidence from genome-wide association studies and replication within multiple myeloma patients. BMC Cancer 18 (1), 820. 10.1186/s12885-018-4728-4

Marsh S., Paul J., King C. R., Gifford G., McLeod H. L., Brown R. (2007). Pharmacogenetic assessment of toxicity and outcome after platinum plus taxane chemotherapy in ovarian cancer: the Scottish randomised trial in ovarian cancer. J. Clin. Oncol. 25 (29), 4528–4535. 10.1200/JCO.2006.10.4752

Martin-Guerrero I., Gutierrez-Camino A., Echebarria-Barona A., Astigarraga I., Garcia de Andoin N., Navajas A., et al. (2019). Variants in vincristine pharmacodynamic genes involved in neurotoxicity at induction phase in the therapy of pediatric acute lymphoblastic leukemia. Pharmacogenomics J. 19 (6), 564–569. 10.1038/s41397-019-0081-5

McLeod H. L., Sargent D. J., Marsh S., Green E. M., King C. R., Fuchs C. S., et al. (2010). Pharmacogenetic predictors of adverse events and response to chemotherapy in metastatic colorectal cancer: results from north American gastrointestinal intergroup trial N9741. JCO 28 (20), 3227–3233. 10.1200/JCO.2009.21.7943

Min Y. G., Lee S.-Y., Lim E., Park M.-Y., Kim D.-h., Byun J. M., et al. (2024). Genetic risk factors for bortezomib-induced neuropathic pain in an Asian population: a genome-wide association study in South Korea. J. Pain 25 (9), 104552. 10.1016/j.jpain.2024.104552

Moore A. S., Norris R., Price G., Nguyen T., Ni M., George R., et al. (2011). Vincristine pharmacodynamics and pharmacogenetics in children with cancer: a limited‐sampling, population modelling approach. J. Paediatr. Child. Health 47 (12), 875–882. 10.1111/j.1440-1754.2011.02103.x

Mufti K., Cordova M., Scott E. N., Trueman J. N., Lovnicki J. M., Loucks C. M., et al. (2024). Genomic variations associated with risk and protection against vincristine-induced peripheral neuropathy in pediatric cancer patients. Npj Genom Med. 9 (1), 56. 10.1038/s41525-024-00443-7

Nakayama H., Ishida H., Iidaka M., Kato S., Nakatani K., Nakayama A., et al. (2024). Association between ABCC2 −24C>T and nab-paclitaxel-induced peripheral neuropathy in Japanese patients with pancreatic cancer. Anticancer Res. 44 (11), 5023–5033. 10.21873/anticanres.17326

Oscanoa J., Sivapalan L., Gadaleta E., Dayem Ullah A. Z., Lemoine Nicholas R., Chelala C. (2020). SNPnexus: a web server for functional annotation of human genome sequence variation (2020 update). Nucleic Acids Res. 48 (W1), W185–W192. 10.1093/nar/gkaa420

Page M. J., McKenzie J. E., Bossuyt P. M., Boutron I., Hoffmann T. C., Mulrow C. D., et al. (2021). The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 372, n71. 10.1136/bmj.n71

Park S. B., Kwok J. B., Asher R., Lee C. K., Beale P., Selle F., et al. (2017). Clinical and genetic predictors of paclitaxel neurotoxicity based on patient- versus clinician-reported incidence and severity of neurotoxicity in the ICON7 trial. Ann. Oncol. 28 (11), 2733–2740. 10.1093/annonc/mdx491

Rizzo R., Spaggiari F., Indelli M., Lelli G., Baricordi O. R., Rimessi P., et al. (2010). Association of CYP1B1 with hypersensitivity induced by taxane therapy in breast cancer patients. Breast Cancer Res. Treat. 124 (2), 593–598. 10.1007/s10549-010-1034-5

Rosenbaum T., Simon S. A. (2007). “TRPV1 receptors and signal transduction,” in TRP Ion Channel Function in Sensory Transduction and Cellular Signaling Cascades, 69–84.

Ruzzo A., Graziano F., Galli F., Giacomini E., Floriani I., Galli F., et al. (2014). Genetic markers for toxicity of adjuvant oxaliplatin and fluoropyrimidines in the phase III TOSCA trial in high-risk colon cancer patients. Sci. Rep. 4 (1), 6828. 10.1038/srep06828

Sałat K. (2020). Chemotherapy-induced peripheral neuropathy: part 1—current state of knowledge and perspectives for pharmacotherapy. Pharmacol. Rep. 72 (3), 486–507. 10.1007/s43440-020-00109-y

Sepe D. M., McWilliams T., Chen J., Kershenbaum A., Zhao H., La M., et al. (2012). Germline genetic variation and treatment response on CCG‐1891. Pediatr. Blood and Cancer 58 (5), 695–700. 10.1002/pbc.23192

Seretny M., Currie G. L., Sena E. S., Ramnarine S., Grant R., MacLeod M. R., et al. (2014). Incidence, prevalence, and predictors of chemotherapy-induced peripheral neuropathy: a systematic review and meta-analysis. Pain 155 (12), 2461–2470. 10.1016/j.pain.2014.09.020

Sherry S. T., Ward M. H., Kholodov M., Baker J., Phan L., Smigielski E. M., et al. (2001). dbSNP: the NCBI database of genetic variation. Nucleic Acids Res. 29 (1), 308–311. 10.1093/nar/29.1.308

Sissung T. M., Mross K., Steinberg S. M., Behringer D., Figg W. D., Sparreboom A., et al. (2006). Association of ABCB1 genotypes with paclitaxel-mediated peripheral neuropathy and neutropenia. Eur. J. Cancer 42 (17), 2893–2896. 10.1016/j.ejca.2006.06.017

Sohani Z. N., Meyre D., de Souza R. J., Joseph P. G., Gandhi M., Dennis B. B., et al. (2015). Assessing the quality of published genetic association studies in meta-analyses: the quality of genetic studies (Q-Genie) tool. BMC Genetics 16 (1), 50. 10.1186/s12863-015-0211-2

Sohani Z. N., Sarma S., Alyass A., de Souza R. J., Robiou-du-Pont S., Li A., et al. (2016). Empirical evaluation of the Q-Genie tool: a protocol for assessment of effectiveness. BMJ Open 6 (6), e010403. 10.1136/bmjopen-2015-010403

Starobova H., Vetter I. (2017). Pathophysiology of chemotherapy-induced peripheral neuropathy. Front. Mol. Neurosci. 10, 174. 10.3389/fnmol.2017.00174

Sucheston-Campbell L. E., Clay-Gilmour A. I., Barlow W. E., Budd G. T., Stram D. O., Haiman C. A., et al. (2018). Genome-wide meta-analyses identifies novel taxane-induced peripheral neuropathy-associated loci. Pharmacogenetics Genomics 28 (2), 49–55. 10.1097/FPC.0000000000000318

Sucheston L. E., Zhao H., Yao S., Zirpoli G., Liu S., Barlow W. E., et al. (2011). Genetic predictors of taxane-induced neurotoxicity in a SWOG phase III intergroup adjuvant breast cancer treatment trial (S0221). Breast Cancer Res. Treat 130, 993–1002. 10.1007/s10549-011-1671-3

Sudlow C., Gallacher J., Allen N., Beral V., Burton P., Danesh J., et al. (2015). UK biobank: an open access resource for identifying the causes of a wide range of complex diseases of middle and old age. PLoS Med. 12 (3), e1001779. 10.1371/journal.pmed.1001779

Tanabe Y., Shiraishi S., Hashimoto K., Ikeda K., Nishizawa D., Hasegawa J., et al. (2020). Taxane-induced sensory peripheral neuropathy is associated with an SCN9A single nucleotide polymorphism in Japanese patients. BMC Cancer 20 (1), 325. 10.1186/s12885-020-06834-0

Tay N., Laakso E. L., Schweitzer D., Endersby R., Vetter I., Starobova H. (2022). Chemotherapy-induced peripheral neuropathy in children and adolescent cancer patients. Front. Mol. Biosci. 9, 1015746. 10.3389/fmolb.2022.1015746

Terrazzino S., Argyriou A. A., Cargnin S., Antonacopoulou A. G., Briani C., Bruna J., et al. (2015). Genetic determinants of chronic oxaliplatin‐induced peripheral neurotoxicity: a genome‐wide study replication and meta‐analysis. J. Peripher. Nerv. Sys 20 (1), 15–23. 10.1111/jns.12110

Van De Velde M. E., Uittenboogaard A., Yang W., Bonten E., Cheng C., Pei D., et al. (2022). Genetic polymorphisms associated with vincristine pharmacokinetics and vincristine-induced peripheral neuropathy in pediatric oncology patients. Cancers 14 (14), 3510. 10.3390/cancers14143510

Vargas-Aliaga A., De La Haba M., Contreras M. J., Morales Estevez C., Porras I., Cano M. T., et al. (2024). NeuroPredict: study of the predictive value of ABCB1 genetic polymorphisms and associated clinical factors in chronic chemotherapy-induced peripheral neuropathy (CIPN). Front. Pharmacol. 15, 1352939. 10.3389/fphar.2024.1352939

Velasco R., Bruna J. (2015). Taxane-induced peripheral neurotoxicity. Toxics 3 (2), 152–169. 10.3390/toxics3020152

Wiernik A., Rogado A., O'Mahony D., Abdul Razak A. R. (2024). Elevating cancer care standards worldwide: an analysis of global initiatives and progress. JCO Glob. Oncol. 10 (10), e2400199. 10.1200/GO.24.00199

Won H. H., Lee J., Park J. O., Park Y. S., Lim H. Y., Kang W. K., et al. (2012). Polymorphic markers associated with severe oxaliplatin‐induced, chronic peripheral neuropathy in colon cancer patients. Cancer 118 (11), 2828–2836. 10.1002/cncr.26614

Wootton R. E., Sallis H. M. (2020). Let’s call it the effect allele: a suggestion for GWAS naming conventions. Int. J. Epidemiol. 49 (5), 1734–1735. 10.1093/ije/dyaa149

Wright G. E., Amstutz U., Drögemöller B. I., Shih J., Rassekh S. R., Hayden M. R., et al. (2019). Pharmacogenomics of vincristine‐induced peripheral neuropathy implicates pharmacokinetic and inherited neuropathy genes. Clin Pharma Ther. 105 (2), 402–410. 10.1002/cpt.1179

Yamada H., Ohmori R., Okada N., Nakamura S., Kagawa K., Fujii S., et al. (2022). A machine learning model using SNPs obtained from a genome-wide association study predicts the onset of vincristine-induced peripheral neuropathy. Pharmacogenomics J. 22 (4), 241–246. 10.1038/s41397-022-00282-8

Yang Y., Zhao B., Gao X., Sun J., Ye J., Li J., et al. (2021). Targeting strategies for oxaliplatin-induced peripheral neuropathy: clinical syndrome, molecular basis, and drug development. J. Exp. and Clin. Cancer Res. 40 (1), 331. 10.1186/s13046-021-02141-z

Yuan Y., Hu W., Chen C., Yao R., Zhang S., Zhu X., et al. (2025). Pharmacokinetic, pharmacodynamic and pharmacogenetic studies related to vincristine-induced peripheral neuropathy in Chinese pediatric ALL patients. Clin. Pharma. Therapeutics 117, 454–464. 10.1002/cpt.3462

Zajączkowska R., Kocot-Kępska M., Leppert W., Wrzosek A., Mika J., Wordliczek J. (2019). Mechanisms of chemotherapy-induced peripheral neuropathy. IJMS 20 (6), 1451. 10.3390/ijms20061451

Zgheib N. K., Ghanem K. M., Tamim H., Aridi C., Shahine R., Tarek N., et al. (2018). Genetic polymorphisms in candidate genes are not associated with increased vincristine-related peripheral neuropathy in Arab children treated for acute childhood leukemia: a single institution study. Pharmacogenetics Genomics 28 (8), 189–195. 10.1097/FPC.0000000000000345

Zhang Y., Zhang H., Wang J., Wei X., Qu Y., Xu F., et al. (2024). A genetic variant study of bortezomib-induced peripheral neuropathy in Chinese multiple myeloma patients. OR 32 (5), 955–963. 10.32604/or.2023.043922

Associated Data

Supplementary Materials

Data Availability Statement

The original contributions presented in the study are included in the article/ Supplementary Material , further inquiries can be directed to the corresponding authors.