Pediatric chemotherapy induced peripheral neuropathy: A systematic review of current knowledge

Accepted Manuscript Pediatric chemotherapy induced peripheral neuropathy: a systematic review of current knowledge Tejaswi Kandula, Susanna B. Park, Richard J. Cohn, Arun V. Krishnan, Michelle A. Farrar PII: DOI: Reference:

S0305-7372(16)30084-6 http://dx.doi.org/10.1016/j.ctrv.2016.09.005 YCTRV 1540

To appear in:

Cancer Treatment Reviews Cancer Treatment Reviews

Received Date: Revised Date: Accepted Date:

19 July 2016 19 August 2016 1 September 2016

Please cite this article as: Kandula, T., Park, S.B., Cohn, R.J., Krishnan, A.V., Farrar, M.A., Pediatric chemotherapy induced peripheral neuropathy: a systematic review of current knowledge, Cancer Treatment Reviews Cancer Treatment Reviews (2016), doi: http://dx.doi.org/10.1016/j.ctrv.2016.09.005

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

PEDIATRIC CHEMOTHERAPY INDUCED PERIPHERAL NEUROPATHY: A SYSTEMATIC REVIEW OF CURRENT KNOWLEDGE RUNNING TITLE: CIPN IN CHILDOOD CANCER Tejaswi Kandulaa,b, Susanna B. Parkc,d, Richard J. Cohna,e, Arun V. Krishnand, Michelle A. Farrara,b a

Discipline of Pediatrics, School of Women’s and Children’s Health, UNSW Medicine, The

University of New South Wales, Randwick, NSW 2031, Australia b

Department of Neurology, Sydney Children’s Hospital, Randwick, NSW 2031, Australia

c

Brain & Mind Centre, University of Sydney, 94-100 Mallett Street, Camperdown, NSW 2050,

Australia d

Prince of Wales Clinical School, UNSW Medicine, The University of New South Wales,

Randwick, NSW 2031, Australia e

Kids Cancer Centre, Department of Oncology, Sydney Children’s Hospital, Randwick,

NSW 2031, Australia

Corresponding author: Dr Michelle Farrar, Department of Pediatric Neurology, Level 4, Emergency Wing, Sydney Children’s Hospital, High Street, Randwick, NSW 2031, Australia T: +61 2 9382 1503 F: +61 2 9382 1580 E-Mail: [email protected] 1

Abstract BACKGROUND:The dramatic increase in the number of childhood cancer survivors over the last 60 years has made monitoring and minimising long term side effects of cancer treatment increasingly important. Chemotherapy induced peripheral neuropathy(CIPN) has been described with many commonly used chemotherapy agents. This article provides a critical overview of pediatric CIPN, its incidence, clinical manifestations, late effects, and recent advances in understanding of risk factors and pharmacogenomics as well as evaluating current assessment strategies and treatment approaches. METHODS:Neurotoxicity data was systematically collated from Medline, Embase and Pubmed and analysed for quality, relevance and originality in three stages prior to inclusion. Quality scoring was done using the QUALSYST assessment tool. RESULTS:A total of 61 studies met inclusion criteria. Peripheral neuropathy is common and may be long lasting with characteristics specific to each chemotherapy agent. There is significant variability in reported incidence and natural history, related to challenges in clinical assessment and diagnosis. Emerging risk factors for CIPN include treatment factors such as dose, duration and concurrent medication and patient factors such as age and inherited susceptibilities. Recent identification of individual genetic variations has advanced understanding of pathomechanisms and may direct future treatment approaches. CONCLUSION:While these studies guide suggestions for current clinical practice, further systematic research with development of strategies for amelioration and prevention of CIPN is necessary. Standardised assessment protocols and objective outcomes measures of CIPN applicable to patients of different ages are critical to enabling the development of novel treatments and facilitation of future clinical trials and treatment individualization.

3

Key words peripheral neuropathy chemotherapy complication chemotherapy induced peripheral neuropathy systematic review cancer survivorship

4

Introduction There has been a steady increase in incidence of childhood cancer diagnoses in developed countries since the 1950s and a concurrent decline in mortality with the overall five-year childhood cancer survival reaching 80% for nearly all cancer types [1]. Accordingly, there is now a burgeoning population of childhood cancer survivors, with an estimated 388,500 long term survivors in the United States alone [2]. It is, therefore, critical to characterise the long lasting side effects of treatment for childhood cancer, as well as accurately assess, monitor and ultimately develop management strategies to prevent persistent treatment related side-effects. Many chemotherapy agents used in cancer treatment can cause acute and chronic peripheral nervous system injury and dysfunction termed ‘chemotherapy induced peripheral neuropathy’ (CIPN). Peripheral nerve toxicity has been described with vinca alkaloids, platinum compounds, taxanes, epothilones, bortezomib and thalidomide. Vinca alkaloids and platinum compounds are the more commonly used agents in childhood cancer and the efficacy and tolerability of several of the other agents is being investigated. Individual chemotherapy agents are known to produce a broadly reproducible pattern of CIPN affecting the sensory, motor and/or autonomic components of the peripheral nervous system [3]. The purpose of this review is to guide current management and future translational research by providing an evidence base of our current understanding of CIPN in children and childhood cancer survivors, pediatric-specific assessment and diagnostic strategies and any available treatment approaches.

Methods

5

Search strategy and selection criteria: This review was undertaken according to the PRISMA guideline for systematic reviews. All published scientific literature on pediatric CIPN was considered for inclusion with the most recent database search performed on 18 February 2016. We searched Medline, Embase and Pubmed for articles using the terms ‘peripheral nervous system diseases’, ‘peripheral nerve injury’, ‘neurotoxicity syndromes’, ‘peripheral neurotoxicity’, ‘peripheral neuropathy’ or ‘toxic neuropathy’ as subject headings or key words, in conjunction with each chemotherapy agent. This was limited to children aged 0 to 18 years or to specific age ranges including preschool children (1 to 6 years), children (6 to 12 years) or adolescents (13 to 18 years). Articles from the authors’ database and bibliographic references cited by original and review articles identified as part of the literature search were also explored and crossreferenced against the search results. An awareness of the relative scarcity of information on pediatric CIPN has led to the design of broad inclusion criteria for this systematic review. The studies were analysed in three stages prior to inclusion in the final review. Case reports, case series, original research articles or systematic reviews published in the English language were appraised in more detail if they had adequate data on CIPN as an assessment or outcome measure, included a minimum of 25% pediatric or adolescent (≤18 years) patients or survivors of childhood cancer in the study population. Animal or basic science research articles, conference abstracts and studies utilising chemotherapy regimens with more than one neurotoxic agent were excluded.

Data extraction and analysis: Original research articles thus identified were scored using the QUALSYST assessment tool [4], a standardised system which gives a score between 0-2 for aspects of study design, 6

selection and description of subject and comparison group characteristics, randomisation and blinding where relevant, description of exposure and outcome measures, sample size, statistical analysis, reporting of results and control for confounding factors. The total score out of a possible 28 is then converted into a standardised score out of 1. Data on the type and topic of the research study, sample size, neurotoxic agent, details and limitations in assessment and reporting of peripheral neurotoxicity as well as the score on each individual QUALSYST item was extracted and recorded in a spreadsheet by TK. A QUALSYST cut-off score of 0.55 was chosen in order to capture 75% of the articles as well as ensure inclusion of several descriptive articles which contained valuable data on clinical characteristics of agent-specific neuropathy. Following quality scoring, the final selection of articles was reviewed by all authors and inclusion was on the basis of originality and relevance (Figure 1 and Appendix A).

Data synthesis: The studies were categorised based on chemotherapy agent and whether they assessed neurotoxicity during acute treatment or long term outcomes. For each agent, further data regarding acute and long term clinical characteristics of neuropathy, electrophysiological changes and pre-disposing risk factors such as age, concomitant administration of other medications and dose-toxicity relationships were collated. The synthesis of the collective data for each agent involved classifying its reliability according to Strength of Recommendation Taxonomy (SORT) criteria (Level A-consistent and good-quality patient-oriented evidence, Level B-inconsistent or limited-quality patientoriented evidence, Level C-consensus, usual practice, opinion, disease-oriented evidence or case series) [5]. In addition, the methods of neurotoxicity assessment that were utilised in

7

these studies have been reported as these are likely to have an impact on the results, particularly in different pediatric age groups.

Results A total of 1580 articles were identified and following removal of duplicates and application of initial filters, 209 records were examined in greater detail. QUALSYST scoring was carried out on 81 research articles that were not excluded for other reasons as detailed in Figure 1. The median QUALSYST score was 0.8 (Range 0.17 – 1.00; IQR 0.65 – 0.89). Case series and case reports were considered separately and not scored using this tool. 61 articles were included in this review for the final qualitative synthesis. This comprised 3 randomised controlled trials, 6 prospective, 9 retrospective and 14 cross-sectional studies, 22 phase I and II clinical trials, 6 case series and 1 case report. These studies form the basis for the following analysis of clinical features, objective assessment and risk factors.

Agent specific peripheral neurotoxicity during treatment: Vincristine neurotoxicity was reported in 52% of the studies while only 5% reported on cisplatin neurotoxicity. 80% of the articles studied neurotoxicity during acute treatment and 20% reported on late effects (Table 1). Details of the individual agents are provided in Table 2 and summarised below. 1. Vinca alkaloids Vincristine treatment produces sensorimotor and autonomic neuropathies. Motor impairment is most prominent in children, manifesting as foot drop, ataxia, gait abnormalities and muscle weakness which can be asymmetric [6-9]. Sensory symptoms include parasthesiae and dysaesthesia. Constipation may occur with autonomic nerve involvement [6-

8

8,10]. There is a universal reduction or absence of deep tendon reflexes [9,11]. Cranial neuropathies characterised by hoarse voice, ptosis and extra-ocular eye movement abnormalities and rarely optic neuropathy have also been reported [12,13]. Vincristine related nerve changes occur early during the course of treatment in children, frequently within the first month [8,11,14]. In a retrospective review of six large treatment trials including a total of 4567 patients [15], severe or disabling peripheral neurotoxicity (National Cancer Institute Common Terminology Criteria for Adverse Events - NCI-CTCAE grade 3 and 4) with vincristine was observed in 10% of patients receiving cumulative doses of 5-10mg/m2, with a significant increase involving 20-52% of patients administered 30mg/m2, suggesting a cumulative dosetoxicity relationship (Table 2). Neuropathy of any grade (NCI-CTCAE 1-4) affected 78100% of patients treated with vincristine and neuropathic pain was seen in up to 35% [8,10,11,14]. Neurophysiological changes occur early in the course of the treatment with conventional nerve conduction studies demonstrating a motor or sensorimotor axonal neuropathy in >90% of patients by the end of 4-5 weeks with weekly administration of vincristine. These changes are motor predominant, may be symmetric or asymmetric and affect the lower limbs and upper limbs, reflecting the clinical presentation [6,7,9]. Motor and sensory evoked potentials also show mild, reproducible prolongation of peripheral nerve latencies [16,17]. 2. Platinum analogues Cisplatin is frequently used in combination therapy with vincristine. The incidence and characteristics of peripheral neuropathy from cisplatin in children are poorly defined with current evidence limited to case reports. 9

Carboplatin is also frequently used in combination therapy with vincristine making it difficult to appreciate its relative contribution to peripheral neurotoxicity. In a single phase II trial of carboplatin in heavily pre-treated patients, peripheral neuropathy was uncommon with World Health Organisation (WHO) grade I and II neuropathy seen in 4% of 93 courses [18]. Oxaliplatin neurotoxicity in pediatric phase I and II trials presented as sensory neuropathy with limb and laryngopharyngeal parasthesia and dysaesthesia, cold related dysaesthesia as well as muscle cramps/spasms and jaw pain [19-22]. In pre-treated patients, acute peripheral neuropathy was a dose limiting toxicity with the maximum tolerated single dose being 130mg/m2 every three weeks and 110mg/m2 with more frequent dosing [19-24]. At 100-130mg/m2 per dose, CTCAE grade 1/2 and grade 3/4 neuropathic symptoms were observed in 37-50% and 3-8% of patients respectively. No clinical cumulative neurotoxicity has been reported with doses above 400 and 1000mg/m2, yet significant reduction in sensory nerve responses on electrophysiological testing, following a minimum cumulative dose of 720mg/m2 was reported in four patients [21]. 3. Microtubule stabilising agents Taxanes: Paclitaxel was trialled as either three, six or 24-hour infusions in pediatric phase I trials, with frequency ranging between twice weekly to once every three weeks [2529]. Peripheral neuropathy was the dose limiting toxicity at single doses of 420mg/m2 presenting as transient parasthesiae in a glove and stocking distribution within one week of the infusion. Transient, severe myalgia was also common [27,29]. Grade 1 or 2 neuropathy was reported in 11-50% and grade 3 or 4 in 6-12% of children treated over an average of 1-2 courses. Cumulative toxicity data is lacking [25-29]. There is some evidence of a dosetoxicity relationship [27]. Smaller, more frequent doses may be associated with reduced neurotoxicity [25].

10

Neuropathy was less common with Docetaxel and not dose-limiting. Grade 3 or 4 sensory neuropathy was observed in 5% of patients, especially with 4 or more courses of treatment [30-33]. The specific characteristics of the neuropathy have not been clearly described. Epothilones: Neuropathy with ixabepilone was predominantly mild, infrequent and non-dose limiting. 1/18 patients (5%) had grade 3 neuropathic pain and 4/18 (22%) had grade 1 sensory neuropathy during the first cycle of the phase I trial [34,35]. 4. Proteasome inhibitors Bortezomib has been utilised to prevent antibody mediated rejection in cardiac and renal transplantation, in addition to phase I and II trials for refractory or relapsed haematological malignancies [36-38]. Grade 3 or 4 peripheral neuropathy was seen in up to 6% and Grade 1 or 2 neuropathy in 10-18% of children receiving 1-2 courses of bortezomib containing chemotherapy [36-38]. Painful sensory neuropathy was observed in one patient, but the nature of the neuropathy in the remainder was not characterised. Off-therapy worsening of symptoms, known as coasting, was documented in one patient [38]. 5. Immunomodulatory drugs Thalidomide has been used in children primarily in the treatment of inflammatory conditions such as refractory inflammatory bowel disease, juvenile rheumatoid arthritis and vasculitis, graft versus host disease and less commonly as part of anti-angiogenic therapy for particular pediatric brain tumours [39-41]. While efficacious, thalidomide neurotoxicity is a dose-limiting side effect and presents with painful distal paresthesias and sensory loss, sometimes associated with proximal muscle weakness and muscle cramps [42]. A subclinical neuropathy may also be observed defined by electrophysiological abnormalities [39,40,42]. In two randomised clinical trials containing 28 11

and 12 children each on thalidomide, clinical peripheral neuropathy requiring dosage reduction or cessation occurs in 20-40% of patients on a dose of 1.5-3mg/kg/day, developing after a time lag of 9-10 months from the beginning of treatment at a minimum cumulative dose of 330-380mg/kg [39,40]. Milder clinical manifestations and/or isolated electrophysiological abnormalities were seen more frequently [39-41]. Neurophysiological studies demonstrate a symmetrical, length-dependent predominantly sensory axonal neuropathy [41,42]. Several cases of off-therapy worsening of neuropathy (coasting) have been described [42]. Improvement or stabilisation in clinical and/or electrophysiological abnormalities was seen in 50% of patients with tapering of the thalidomide dose [39].

Long term outcomes in childhood cancer survivors With increasing survival, focus is now shifting to medium and long term outcomes with emerging long term CIPN data for vincristine, cisplatin and thalidomide. Clinical and electrophysiological: In children with acute lymphoblastic leukaemia (ALL) treated with multiple cycles of vincristine, objective electrophysiological evidence of persistent motor or sensorimotor axonal neuropathy was evident in 30-33% of children up to seven years following treatment suggesting that vincristine related peripheral nerve changes in children can be long lasting [43-46]. An additional 27% reported subjective symptoms of neuropathy without abnormalities on nerve conduction studies. 10 out of 29 (35%) of adolescents and young adults tested a median of 8 months after completion of their cisplatin therapy demonstrated a sensory axonal neuropathy with clinical abnormalities in vibration perception threshold and deep tendon reflexes. The presence of abnormalities on nerve conduction studies and the degree of impairment in vibration perception correlated with the cumulative dose of cisplatin (>300mg/m2) [47]. 12

Persistent clinical and electrophysiological deficits following thalidomide, continued to be observed in a proportion of children up to 6-9 months after cessation of treatment [41,42]. Functional: St Jude’s lifetime cohort studies demonstrated sensory and motor impairment in 20% and 17.5%, respectively, of 531 adult survivors of childhood extra-cranial solid tumours [48,49]. Among 415 adult survivors of ALL poorer balance, mobility and 6minute walk distances were seen in 15.4%, 3.6% and 46.5% of participants. Vinca-alkaloid exposure was associated with increased risk of motor impairment and platinum exposure was associated with increased risk of sensory impairment [48,49]. These findings were reproducible, with survivors of childhood cancer having reduced performance in motor tasks, peripheral strength, motor dexterity, balance scores and decreased preference for physical activity compared to controls [44,50-53]. No consistent association has been demonstrated between doses of vincristine and degree of long term impairment.

Assessment techniques Studies assessing CIPN have utilised various standardised assessment techniques as a means of quantifying degree of impairment (Table 3). Of the more commonly used tools, the National Cancer Institute Common Terminology Criteria for Adverse Events (NCI-CTCAE) [54] was a clinical grading scale employed in 41%, the Total Neuropathy Score (TNS) [55-57] in 11.5% and the Modified Balis Pediatric Scale [58,67] in 8.2% of the studies. The NCI-CTCAE scale is graded from 1-5 for motor and sensory neuropathy based on neuropathic symptoms and impact on activities of daily living [54]. The Modified Balis Pediatric Scale is a modified version of the CTCAE score and incorporates greater detail about symptoms, severity and activities specific to children [67]. The Total Neuropathy score is a composite scale with motor, sensory and autonomic symptom report and multimodality examination findings each scored from 0-4 [56] 13

(Appendix B). Other grading scales included the World Health Organisation toxicity grading scales, FACES pain scale, Neuropathy Impairment Scale and the Michigan autonomic symptoms survey [18,28,44,57]. Specific clinical grading scales designed to incorporate the unique characteristics of oxaliplatin neuropathy were used when relevant [19-22]. Composite functional measures such as Bruininks-Oseretsky Test of Motor Proficiency and the Movement Assessment Battery for Children as well as individual measures such as the purdue peg-board, handwriting assessment, quantitative sensory testing, muscle strength, range of movement and sensory organisation (balance) testing, timed up and go and sixminute walk tests have been used as surrogate measures for physical function [35,44,4853,59]. Nerve conduction studies were utilised in 24.6% of the studies. Other electrophysiological measures included motor and sensory evoked potentials and autonomic function tests [6,7,16,43-45]. Quality of life measures were seldom used and the Pediatric Quality of Life inventory (PedsQL) was used in a single study [44]. There are no pediatric CIPN specific patient reported outcome (PRO) measures in use.

CIPN Risk factors Studies have observed that a subgroup of children experience a greater severity of CIPN [11], which may be related to several risk factors. 1. Treatment factors: Dose - Greater peripheral neurotoxicity has been observed in children receiving higher cumulative doses of vincristine [15] and preliminary data suggests that a dose-severity relationship may also be evident with other chemotherapeutic agents (Table 2).

14

Duration – Thalidomide neuropathy in patients receiving 1.5-3mg/kg/day was seen to develop 9-10 months from the beginning of treatment [39-41]. No neuropathy was seen in patients receiving high daily doses (12mg/kg/day) for less than three months suggesting that duration of treatment rather than the single dose may be important [60]. Concurrent medication – concomitant administration of triazole or imidazole antifungal agents with vincristine is associated with increased frequency and severity of peripheral neurotoxicity [61,62]. Median time lag to manifestation of toxicity may be up to 30 days and varies with the azole antifungal agent used [61,62]. A single study has reported the incidence of several cases of severe atypical neuropathy with administration of haematopoietic colony stimulating factors in association with vincristine chemotherapy [63]. Case reports of pharmacokinetic interactions with other drugs include nifedipine, cyclosporin, carbamazepine and phenytoin. Level of evidence for dose and duration of treatment as risk factors for neuropathy is dependent on the specific neurotoxic agent (Table 2), whereas the evidence for concurrent medication is based on several case series and is classed as level C [5]. 2. Disease factors: Haematological malignancy - A severe, rapid onset, progressive weakness, resembling Guillain Barré Syndrome, has been described in children with haematological malignancies prescribed vincristine, usually occurring during the induction phase of chemotherapy [64,65]. The level of evidence for disease related risk factors is level C [5]. 3. Patient factors:

15

Age - A progressive increase in risk of vincristine neurotoxicity with older age has been observed in several studies including large treatment trials [11,15,62]. Similarly, with thalidomide older age was associated with more frequent neuropathy in one case series [41]. Genetic alterations in pharmacokinetic pathways (Figure 2) - It has been observed that ‘low-expressors’ of the cytochrome P450 enzyme CYP3A5, have a greater incidence and severity of vincristine neurotoxicity than their ‘high-expressor’ counterparts [14]. Genetic polymorphism in another pharmacokinetic gene, ABCB1, was also noted to alter risk of neurotoxicity [66]. Genetic alterations in pharmacodynamic pathways - A recent large genome wide association study established allelic variation of the CEP72 gene, involved in microtubule formation, as being significantly associated with vincristine neuropathy in children [67]. A further study that investigated polymorphisms in several pharmacodynamic genes (TUBB1, MAP4, ACTG1 and CAPG) also found allelic variations that may alter the risk of neuropathy [66]. Genetic susceptibility to hereditary neuropathy - Numerous cases of children presenting with unexpected severe chemotherapy induced neurotoxicity have been reported, subsequently diagnosed as having a previously unrecognised inherited neuropathy [68]. The level of evidence for patient related risk factors is level B for age and level B-C for genetic risk factors [5].

Therapeutic options - Prevention and treatment There has been very limited examination of prevention and therapeutic approaches in pediatric CIPN.

16

A single phase I study investigating oxaliplatin therapy, in patients aged 5 to 21 years, described the use of carbamazepine to prevent dose-limiting peripheral neurotoxicity, allowing escalation of dose beyond the previous maximum tolerated dose in 6 patients [22]. Utilising an oxaliplatin-specific sensory neuropathy scale designed by the drug manufacturer, it was observed that while acute oxaliplatin-induced sensory neuropathy was common in the carbamazepine treated group, it was not severe enough to be dose-limiting. While a pilot single-blinded study of oral glutamic acid as a protective agent against vincristine toxicity showed promise [69], a further larger double-blinded randomised controlled trial did not replicate these findings [58]. Both these studies used clinical grading scales to evaluate vincristine neuropathy incorporating symptom report and examination findings of sensory, motor and autonomic manifestations in children with a mean age of 7-9 years. Amelioration of vincristine induced neurotoxicity with pyridoxine and pyridostigmine has been observed in isolated cases [70]. One case series reports four patients, 2-13 years old, presenting with loss of ambulation or complete paralysis of lower limbs 1-7 days after the fourth dose of vincristine (cumulative dose 3-8mg), which improved within 1-2 weeks of administration of pyridoxine/pyridostigmine. Vincristine was withheld in three of the patients and continued as per protocol in the fourth [70]. The use of gabapentin in vincristine treated patients has been documented in some retrospective studies, but not systematically studied [10,61]. Neuropathic pain was graded using the NCI-CTCAE criteria but there is absence of consistent severity data before and after administration of gabapentin [10]. The use of intravenous immunoglobulin as a successful therapeutic intervention has been reported in case series and reports of patients with haematological malignancy in the specific context of rapid onset, severe neuropathy

17

resembling Guillain Barré Syndrome (clinical and electrophysiological), sometimes allowing the continuation of the chemotherapy agent [64,65]. This has, however, not be studied systematically and improvement frequently coincides with remission of disease. Accordingly, there is only level C evidence [5] for preventative and therapeutic interventions in CIPN, such that caution needs to be exercised in clinical implementation.

Discussion The rapidly growing population of childhood cancer survivors means that there is a large cohort of patients potentially at risk of long term sequelae from CIPN. Furthermore, neuropathy can be dose-limiting, thus restricting the use of the chemotherapy and finding strategies to overcome this is important. Long term motor impairment may contribute towards reduced preference for physical activity and life style choices which compound metabolic disorders such as diabetes, obesity and osteoporosis. While the physical limitations seen in childhood cancer survivors are multifactorial, it is important to further understand the role of CIPN as a contributing factor to these functional impairments in order to direct appropriate intervention. Optimisation of cancer chemotherapy requires treatment individualisation not only to maximise survival, but also to minimise toxicity. To this end, the current knowledge and the need for future research into CIPN can be divided into a few distinct themes.

Understanding the manifestations, incidence and natural history of agent-specific neuropathy This will aid anticipation and accurate diagnosis of neuropathic symptoms in younger children less able to articulate their symptoms. The diagnostic process is complicated by confounding factors such as systemic illness and concurrent use of steroids. Incidence,

18

characteristics and natural history data of agent-specific CIPN in children is emerging from retrospective, cross-sectional and a few prospective studies for vincristine as well as newer agents. There are, however, significant gaps in our knowledge, particularly for commonly used agents such as cisplatin and carboplatin. In addition, observations such as the motor predominance of vincristine neuropathy in children, compared to the sensory predominance in adults3 highlight the difficulties with extrapolating adult data to pediatric situations. A further limitation is that the majority of early clinical trials for newer agents administered the chemotherapy agent for 1-2 cycles to individuals who were pre-treated with other neurotoxic agents. This limits the capacity to interpret both acute and cumulative neurotoxicity. Taken together, further systematic research is required to gain a better understanding of CIPN in different pediatric populations with each chemotherapeutic agent. Encouragingly, more specific and systematic assessments to identify neurotoxicity have been employed in recent oxaliplatin, ixabepilone and thalidomide clinical trials thus maximising the accuracy of the data collected [21,35,39,40].

Development of age-appropriate, standardised assessment protocols The predominant limitation of many of the studies reviewed is the variability in the use of age-appropriate, validated neurotoxicity assessments (Table 3), both for diagnosis and ascertainment of severity. Standardised assessment protocols for CIPN are critical to understanding the onset and development of neurotoxicity, track progression, enabling accurate comparison between clinical trials and the development of neuroprotective approaches. Underscoring this concept, the proportion of children experiencing significant vincristine toxicity ranged from 1% to 52% in large clinical trials [15]. In addition to variance in vincristine dosing and treatment protocols, this wide variability may be attributable to the retrospective nature of many studies, differences in reporting criteria, differences in

19

ascertainment of toxicity data between phase I and phase III studies and difficulties in recognition of CIPN in children. Clinical manifestations are often non-specific and assessment needs to be tailored to the age of the child (Appendix C). The most commonly used standardised clinical grading scale in clinical trials for all age groups, the NCI-CTCAE scale, is easy to use and has been shown to have good inter and intra-rater reliability in adults, provided assessors are trained for consistency of measurement [71]. However, it relies on symptom report, has limited sensitivity to change and may be prone to ambiguity in interpreting the functional impact of neuropathic symptoms in children of different ages leading to inaccuracy and under reporting of neuropathy in this population [55,57]. The Modified Balis Pediatric Scale incorporates greater detail regarding symptoms and functional tasks specific to children, but has also been demonstrated to have limited sensitivity to change [57]. The Total Neuropathy Score, a composite score for the assessment of CIPN incorporating peripheral neurological symptoms and examination findings, also has good inter and intra-rater reliability and has been shown to have greater sensitivity to change [55,57,71]. A validated and reliable pediatric version of the TNS score, the Pedm-TNS, and a vincristine specific version, TNS-PV, are available for use in children between 5-18 years of age (Table 3) [55-57]. These clinical grading scales may need to be combined with CIPN-specific functional assessments, quality of life measures as well as objective electrophysiological techniques in order to gain an appreciation of the acute and long term impact of CIPN. Development and incorporation of patient reported outcome measures is important to augment clinician based assessment tools which can underestimate the impact of CIPN [72]. In the absence of pediatric CIPN specific PRO measures, those developed for use in other contexts like musculoskeletal disorders incorporating upper and lower limb function, self-care and mobility items such as the Pediatric Outcomes Data Collection Instrument (PODCI) and Activities Scale for Kids (ASK) may be 20

useful [73]. In addition, adult CIPN functional PRO scales with good validity and reliability scores such as the European Organization for Research and Treatment of Cancer (EORTC) CIPN20 and the Functional Assessment of Cancer Therapy/Gynaecologic Oncology Group-neurotoxicity (FACT/GOG-NTx) subscale may be able to be validated and utilised in the adolescent and young adult population [71,74].

Identifying risk factors that predispose to CIPN Individual differences in sensitivity to CIPN, a phenomenon frequently seen in clinical practice, may be due to multiple susceptibility or protective factors. Understanding these factors is a crucial step towards treatment individualisation. Determination of risk may need to include factors such as disease characteristics, type of chemotherapy agent, dosing schedules, infusion durations, combination regimes, age, inherited susceptibilities and the concurrent administration of certain medications. Older age has been found by many studies to be associated with increased incidence of neuropathy for vincristine and thalidomide, although the reason for this observation is unclear and may be related to limitations in CIPN assessment in different age groups. Single and cumulative dose-toxicity relationships are starting to be understood for some agents and further data is required with other agents. Coadministration of certain medications, especially triazole and imidazole anti-fungal agents, is thought to exacerbate vincristine induced toxicity through pharmacokinetic inhibition of the cytochrome P450 and P-glycoprotein mediated vincristine metabolism and transport [61,62]. An appreciation of each of these risk factors allows the clinician to tailor therapy to the unique situation of each individual.

Pharmacogenetic advances Recent advances in genetic sequencing have allowed exploration of multiple single base pair genetic mutations or nucleotide polymorphisms (SNP) which may alter 21

chemotherapy neurotoxicity profiles. The majority of the pediatric research exploring genetic susceptibilities has been in vincristine induced peripheral neuropathy. The candidate SNPs can be broadly divided into three categories - those affecting vincristine metabolism (pharmacokinetics), those affecting the vincristine toxicity pathways (pharmacodynamics) and SNPs in hereditary neuropathy genes (Figure 2). The cytochrome P450 enzyme CYP3A5 plays a key role in vincristine metabolism. Allelic variation in the gene for CYP3A5 results in phenotypic differences in the expression of functional enzyme. This may be the basis for the differences in observed rates of vincristine toxicity between Caucasian children and their African-American counterparts [14]. SNPs in microtubule and cytoskeletal genes, the pharmacodynamic target of vincristine, may also alter CIPN risk such. The CEP72 gene, involved in microtubule formation, identified as a risk factor through genome wide association studies, would fulfil the criteria for biological plausibility [67]. This finding was reproducible in two independent cohorts of patients and the genotype had a significant correlation with the onset-time and severity of peripheral neuropathy, strengthening the argument for this being a significant polymorphism. Many other pharmacokinetic and pharmcodynamic genes have, however, been explored and the findings require replication in other patient cohorts [66]. The results have been mixed, highlighting the importance of adequate sample size and precise definition of peripheral neuropathy and its severity. The third group of mutations are those in hereditary neuropathy genes. Early and severe vincristine induced peripheral neuropathy can occur in children with an inherited underlying susceptibility in the form of a subclinical or manifest hereditary neuropathy such as CharcotMarie-Tooth disease. Approximately 80 genes have been characterised related to CharcotMarie-Tooth disease and genetic variation in these genes is over represented in adult patients with susceptibility to CIPN [75]. This requires further assessment in children. 22

Diagnosis and treatment Identification of patients vulnerable to severe CIPN through baseline neurological assessment and consideration of recognised risk factors and family history is important. In addition, early diagnosis through routine screening for manifestations of CIPN will enable timely intervention. A simple, standardised clinical approach to a child presenting with possible CIPN to direct subsequent investigation is critical and an algorithm based on current understanding of agent specific neuropathy is presented (Figure 3 and Appendix C). Routine neurophysiological assessment is difficult in young children due to the associated discomfort. As such, further investigation may be indicated if the clinical presentation is atypical of the currently recognised manifestations of agent-specific peripheral neuropathy (Figure 3). Nerve conduction studies are a useful adjunct in children both to determine the pattern of neuropathic involvement but also to assist with the differential diagnosis which includes vitamin deficiencies (Vit B12, Vit E), Guillain Barré Syndrome, critical illness neuropathy, inherited neuropathies, steroid myopathy, infection and nerve root infiltration [64,68,76]. There are currently no evidence based preventative or treatment strategies for CIPN in children. A recent comprehensive review of multiple strategies in adults, which forms the basis for the American Society of Clinical Oncology (ASCO) guidelines, has reported a lack of high quality, consistent data to support the use of agents other than duloxetine [77]. Of the many potential neuroprotective agents that have been used in adults, the only ones that have been trialled in children are carbamazepine and glutamic acid for the prevention of CIPN, and intravenous immunoglobulin, pyridoxine/pyridostigmine and gabapentin for treatment, with limited evidence (Level C) for benefit [10,12,22,58,64,70]. Some of the agents are reported in case series or retrospective studies making it difficult to differentiate treatment effect from the expected natural history of the neuropathy

23

Agents used in routine clinical practice are those that are efficacious in other neuropathic conditions including tricyclic antidepressants, GABAergic agents and topical preparations [77]. Modification of dosing schedules for significant clinical neuropathy (NCICTCAE grade 3 or 4) remains the mainstay of management, although there is no direct evidence to support this as an efficacious strategy. While a comprehensive appraisal of the merits of early rehabilitation or exercise in CIPN has not been undertaken, physiotherapy and occupational therapy to minimise limitation in range of movement, deconditioning and long term motor impairment, and exercise to maintain strength and function, as in other neuropathic conditions, would generally be considered important [78].

Future directions The translational approach for treating and preventing CIPN, while maximising the efficacy of chemotherapy agents, includes further understanding of its characteristics, natural history, pathomechanisms and risk factors and the development and validation of standardised and age appropriate assessment protocols. As we move towards better chemotherapy regimens and increased use of new generation agents in children, it is imperative that further research into pediatric CIPN is undertaken in order to optimise quality of life and productivity for long term survivors of childhood cancer.

24

Table 1 – Agent specific CIPN literature reporting on acute toxicity during treatment and long term outcomes

Acute toxicity

Long term outcome

Neurotoxic Agent

Median QUALSYST score*

Number of studies

References

Median QUALSYST score*

Number of studies

References

Vincristine

0.74

21

6-12,14, 15,17,57,58, 61-68,70

0.86

9

16,43-46, 48,51-53

-

0

-

0.82

1

47

Carboplatin

0.64

1

18

-

0

-

Oxaliplatin

0.90

6

19-24

-

0

-

Paclitaxel

0.83

5

25-29

-

0

-

Docetaxel

0.85

4

30-33

-

0

-

Ixabepilone

0.75

2

34,35

-

0

-

Bortezomib

0.83

3

36-38

-

0

-

Thalidomide

0.87

5

39-42,60

-

0

-

More than one agent

0.82

2

55,56

0.93

2

49,50

Cisplatin

*QUALSYST scoring on the basis of the overall quality of the study, rather than the quality of the neurotoxicity assessment Study quality: score of 0.85-1 = very good, 0.7-0.85 = good, 0.55-0.7 = adequate

25

Table 2 – Neurotoxicity from specific chemotherapeutic agents Chemotherapy agent

Utilisation in children

Predominant type of neurotoxicity*

NCS/EMG findings

Incidence**

Vincristine

Cancer treatment

Sensorimotor and autonomic

Motor or sensorimotor axonal neuropathy

Any grade – 78100% Grade 3/4 – 1052% depending on cumulative dose

Prominent motor features in children Cranial neuropathy

Acute/chronic neurogenic changes on EMG

Evidence for dose-toxicity relationship Increased toxicity with greater cumulative dose

Level of evidence# and Potential for bias Level B Different methods of ascertainment of neurotoxicity between clinical trials may affect reported incidence

Ref

611,14,1 5

Can be asymmetric

Cisplatin

Cancer treatment

Sensory neuropathy

Carboplatin

Cancer treatment

Oxaliplatin

Under evaluation for cancer treatment

Sensory neuropathy Acute sensory neuropathy with cold exacerbation and muscle cramps

Paclitaxel

Under evaluation for cancer treatment

Chronic sensory neuropathy Sensory neuropathy

Affects both upper and lower limbs NA

NA Sensory axonal neuropathy

NA

Motor/autonomic involvement less common Docetaxel

Ixabepilone

Bortezomib

Thalidomide

Under evaluation for cancer treatment

Under evaluation for cancer treatment

Sensory neuropathy Motor/autonomic involvement less common Sensory neuropathy

Under evaluation for cancer treatment and organ transplantation

Painful sensory neuropathy

Cancer treatment GVHD Inflammatory bowel disease Rheumatological conditions

Sensory neuropathy

NA

NA

NA

Unclear – mainly case reports NA

NA

NA

NA

NA

NA

NA

Grade 1/2 – 3750% Grade 3/4 - 38%

Increased toxicity with greater single dose

Level C

19-24

Grade 1/2 – 1150% Grade 3/4 - 612%

MTD 130mg/m2 Increased toxicity with greater single dose

Grade 3/4 – 5%

MTD 420mg/m2 Increased toxicity with greater cumulative dose

Grade1/2 – 22% Grade 3/4 - 25%

NA

Grade 1/2 – 1018% Grade 3/4 – 6%

NA

Grade 2 or greater – 2040%

Increased toxicity with greater cumulative dose/greater duration of exposure

Motor/autonomic involvement less common

Motor involvement less common

Symmetrical, length dependent sensory axonal neuropathy Motor axonal neuropathy less pronounced

All patients pre-treated Majority received 2 cycles – limited data on cumulative toxicity Level C

25-29

All patients pre-treated Majority received 1-2 cycles – no data on cumulative toxicity Method of ascertainment may affect neurotoxicity incidence Level C

30-33

All patients pre-treated Method of ascertainment may affect neurotoxicity incidence Level C

34,35

All patients pre-treated Majority received 1-2 cycles – no data on cumulative toxicity Level C

36-38

All patients pre-treated Majority received 1-2 cycles – no data on cumulative toxicity Method of ascertainment may affect neurotoxicity incidence Level A

39-42

Most research done in the context of inflammatory bowel disease and cancer treatment dosing schedules may be substantially different

Mild acute/chronic neurogenic changes on EMG *From pediatric and adult data; **Grade 1-4 based on the National Cancer Institute Common Terminology Criteria for Adverse Events scoring system; # Level of evidence according to Strength of Recommendation Taxonomy (SORT) criteria – Level A- consistent and good quality patient-oriented evidence, Level B-

26

inconsistent or limited-quality patient-oriented evidence, Level C-consensus, usual practice, opinion, disease-oriented evidence or case series5; NCS/EMG – Nerve Conduction Studies/Electromyogram, NA – Not available, MTD – maximum tolerated dose, GVHD – Graft versus host disease

27

Table 3 – Assessment scales/techniques commonly used by CIPN studies Assessment category*

Standardised assessment

Number of studies (%)

Age range of tool use

Pediatric specific scale/protocol available?

Need for assessor expertise

Strengths/weaknesses55,57

Clinical grading scales

NCI-CTCAE

25 (41)

All

No

Moderate

Total Neuropathy Score**

7 (11.5)

5-18yrs

Yes

ModerateHigh

Modified Balis Pediatric scale

5 (8.2)

All Pediatric

Yes

Moderate

FACES pain scale

3 (4.9)

2-18yrs

Yes

LowModerate

Bruininks-Oseretsky Test of Motor Proficiency Movement assessment battery for children Nerve conduction studies

3 (4.9)

4-21yrs

Yes

High

Administration time <10min; Limited sensitivity to change Administration time <10min; Good sensitivity to change Administration time <10min; Limited sensitivity to change Administration time <5min; Limited to pain manifestations Time consuming; Utilised but not validated in CIPN

2 (3.3)

3-16yrs

Yes

High

15 (24.6)

All

Yes

High

Motor/Sensory evoked potentials

4 (6.6)

All

Yes

High

PedsQL

1 (1.6)

All

Yes

Low

Functional scales

Electrophysiological studies

Quality of life measures

Time consuming; Utilised but not validated in CIPN Gold standard objective assessment of peripheral nerves; Time consuming; Associated with patient discomfort; Time consuming; Limited assessment of peripheral nerves Administration time <10min; Not specific to CIPN

NCI-CTCAE - National Cancer Institute Common Terminology Criteria for Adverse Events; *Most commonly used scale/techniques in each category have been included; **Different versions of Total Neuropathy Score (TNS) including modified pediatric TNS, vincristine specific TNS and TNS reduced.

28

Figure Legends Figure 1: Study flow chart; *Inadequate data = inadequate assessment or reporting of peripheral neuropathy/ lack of distinction between central and peripheral neurotoxicity/ <25% pediatric patients/ inadequate separation of adult and pediatric data/ dual neurotoxic agents; **includes case series/reports not scored using QUALSYST; ***Based on originality and relevance to topic Figure 2: Mechanisms of genetic susceptibility to CIPN – Inherited polymorphisms may be in (A) Pharmacokinetic pathways causing alterations in drug metabolism (B) Pharmacodynamic pathways/Neurotoxicity targets such as microtubules or (C) Peripheral neuropathy genes such as those associated with Charcot Marie Tooth disease Figure 3: Proposed algorithm for further assessment of children with possible/suspected CIPN based on differential diagnosis and deviation from currently recognised pattern and severity of patient symptoms and signs; Pattern of neuropathic involvement on nerve conduction studies may assist with differentiating CIPN from other possible diagnoses.

29

References [1] Kaatsch P. Epidemiology of childhood cancer. Cancer treatment reviews. 2010;36:277-85. [2] Phillips SM, Padgett LS, Leisenring WM, Stratton KK, Bishop K, Krull KR, et al. Survivors of childhood cancer in the United States: prevalence and burden of morbidity. Cancer epidemiology, biomarkers & prevention : a publication of the American Association for Cancer Research, cosponsored by the American Society of Preventive Oncology. 2015;24:653-63. [3] Park SB, Goldstein D, Krishnan AV, Lin CS, Friedlander ML, Cassidy J, et al. Chemotherapyinduced peripheral neurotoxicity: A critical analysis. CA: a cancer journal for clinicians. 2013;63:419–437 [4] Kmet LM, Lee RC, Cook LS. Standard quality assessment criteria for evaluating primary research papers from a variety of fields. Alberta Heritage Foundation for Medical Research2004. [5] Ebell MH, Siwek J, Weiss BD, Woolf SH, Susman J, Ewigman B, et al. Strength of recommendation taxonomy (SORT): a patient-centered approach to grading evidence in the medical literature. American family physician. 2004;69:548-56. [6] Courtemanche H, Magot A, Ollivier Y, Rialland F, Leclair-Visonneau L, Fayet G, et al. Vincristine-induced neuropathy: Atypical electrophysiological patterns in children. Muscle & nerve. 2015;52:981-5 [7] Purser MJ, Johnston DL, McMillan HJ. Chemotherapy-induced Peripheral Neuropathy Among Paediatric Oncology Patients. The Canadian journal of neurological sciences Le journal canadien des sciences neurologiques. 2014;41:442-7. [8] Arzanian MT, Mehdizadeh M, Zamani GR. Vincristine induced neurotoxicity: Study of 75 cases. Iranian Journal of Child Neurology. 2009;3:39-44. [9] Toopchizadeh V, Barzegar M, Rezamand A, Feiz AH. Electrophysiological consequences of vincristine contained chemotherapy in children: A cohort study. Journal of Pediatric Neurology. 2009;7:351-6.

30

[10] Anghelescu DL, Faughnan LG, Jeha S, Relling MV, Hinds PS, Sandlund JT, et al. Neuropathic pain during treatment for childhood acute lymphoblastic leukemia. Pediatric blood & cancer. 2011;57:1147-53. [11] Lavoie Smith EM, Li L. Patterns and severity of vincristine-induced peripheral neuropathy in children with acute lymphoblastic leukemia. 2015;20:37-46. [12] Bay A, Yilmaz C, Yilmaz N, Oner AF. Vincristine induced cranial polyneuropathy. Indian J Pediatr. 2006;73:531-3. [13] Teichmann KD, Dabbagh N. Severe visual loss after a single dose of vincristine in a patient with spinal cord astrocytoma. J Ocul Pharmacol. 1988;4:117-21. [14] Egbelakin A, Ferguson MJ, MacGill EA, Lehmann AS, Topletz AR, Quinney SK, et al. Increased risk of vincristine neurotoxicity associated with low CYP3A5 expression genotype in children with acute lymphoblastic leukemia. Pediatric blood & cancer. 2011;56:361-7. [15] Langholz B, Skolnik JM, Barrett JS, Renbarger J, Seibel NL, Zajicek A, et al. Dactinomycin and vincristine toxicity in the treatment of childhood cancer: a retrospective study from the Children's Oncology Group. Pediatric blood & cancer. 2011;57:252-7. [16] Harila-Saari AH, Huuskonen UE, Tolonen U, Vainionpaa LK, Lanning BM. Motor nervous pathway function is impaired after treatment of childhood acute lymphoblastic leukemia: a study with motor evoked potentials. Medical and pediatric oncology. 2001;36:345-51. [17] Vainionpaa L, Kovala T, Tolonen U, Lanning M. Vincristine therapy for children with acute lymphoblastic leukemia impairs conduction in the entire peripheral nerve. Pediatric neurology. 1995;13:314-8. [18] Fagundes Loss J, Albino Santos PP, Di Leone L, Lunardi Brunetto A. Outcome of Pediatric Recurrent and Refractory Malignant Solid Tumors Following Ifosfamide/Carboplatin/Etoposide (ICE): A Phase II Study in a Pediatric Oncology Centre in Brazil. Pediatric Blood and Cancer. 2004;42:139-44. [19] Beaty O, Berg S, Blaney S, Malogolowkin M, Krailo M, Knight R, et al. A phase II trial and pharmacokinetic study of oxaliplatin in children with refractory solid tumors: a Children's Oncology Group study. Pediatric blood & cancer. 2010;55:440-5. 31

[20] Geoerger B, Chisholm J, Le Deley MC, Gentet JC, Zwaan CM, Dias N, et al. Phase II study of gemcitabine combined with oxaliplatin in relapsed or refractory paediatric solid malignancies: An innovative therapy for children with Cancer European Consortium Study. Eur J Cancer. 2011;47:2308. [21] Geoerger B, Doz F, Gentet JC, Mayer M, Landman-Parker J, Pichon F, et al. Phase I study of weekly oxaliplatin in relapsed or refractory pediatric solid malignancies. Journal of Clinical Oncology. 2008;26:4394-400. [22] Spunt SL, Freeman BB, 3rd, Billups CA, McPherson V, Khan RB, Pratt CB, et al. Phase I clinical trial of oxaliplatin in children and adolescents with refractory solid tumors. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2007;25:2274-80. [23] Fouladi M, Blaney SM, Poussaint TY, Freeman BB, 3rd, McLendon R, Fuller C, et al. Phase II study of oxaliplatin in children with recurrent or refractory medulloblastoma, supratentorial primitive neuroectodermal tumors, and atypical teratoid rhabdoid tumors: a pediatric brain tumor consortium study. Cancer. 2006;107:2291-7. [24] Macy ME, Duncan T, Whitlock J, Hunger SP, Boklan J, Narendren A, et al. A multi-center phase Ib study of oxaliplatin (NSC#266046) in combination with fluorouracil and leucovorin in pediatric patients with advanced solid tumors. Pediatric blood & cancer. 2013;60:230-6. [25] Hayashi RJ, Blaney S, Sullivan J, Weitman S, Vietti T, Bernstein ML, et al. Phase 1 study of Paclitaxel administered twice weekly to children with refractory solid tumors: a pediatric oncology group study. Journal of pediatric hematology/oncology. 2003;25:539-42. [26] Horton TM, Ames MM, Reid JM, Krailo MD, Pendergrass T, Mosher R, et al. A Phase 1 and pharmacokinetic clinical trial of paclitaxel for the treatment of refractory leukemia in children: a Children's Oncology Group study. Pediatric blood & cancer. 2008;50:788-92. [27] Hurwitz CA, Relling MV, Weitman SD, Ravindranath Y, Vietti TJ, Strother DR, et al. Phase I trial of paclitaxel in children with refractory solid tumors: a Pediatric Oncology Group Study. Journal of Clinical Oncology. 1993;11:2324-9.

32

[28] Doz F, Gentet JC, Pein F, Frappaz D, Chastagner P, Moretti S, et al. Phase I trial and pharmacological study of a 3-hour paclitaxel infusion in children with refractory solid tumours: a SFOP study. Br J Cancer. 2001;84:604-10. [29] Geller JI, Wall D, Perentesis J, Blaney SM, Bernstein M, Pediatric Oncology Group s. Phase I study of paclitaxel with standard dose ifosfamide in children with refractory solid tumors: a Pediatric Oncology Group study (POG 9376). Pediatric blood & cancer. 2009;52:346-50. [30] Blaney SM, Seibel NL, O'Brien M, Reaman GH, Berg SL, Adamson PC, et al. Phase I trial of docetaxel administered as a 1-hour infusion in children with refractory solid tumors: a collaborative pediatric branch, National Cancer Institute and Children's Cancer Group trial. Journal of Clinical Oncology. 1997;15:1538-43. [31] Seibel NL, Blaney SM, O'Brien M, Krailo M, Hutchinson R, Mosher RB, et al. Phase I trial of docetaxel with filgrastim support in pediatric patients with refractory solid tumors: a collaborative Pediatric Oncology Branch, National Cancer Institute and Children's Cancer Group trial. Clinical Cancer Research. 1999;5:733-7. [32] Yoon JH, Kwon MM, Park HJ, Park SY, Lim KY, Joo J, et al. A study of docetaxel and irinotecan in children and young adults with recurrent or refractory Ewing sarcoma family of tumors. BMC Cancer. 2014;14:622. [33] Zwerdling T, Krailo M, Monteleone P, Byrd R, Sato J, Dunaway R, et al. Phase II investigation of docetaxel in pediatric patients with recurrent solid tumors: A report from the children's oncology group. Cancer. 2006;106:1821-8. [34] Jacobs S, Fox E, Krailo M, Hartley G, Navid F, Wexler L, et al. Phase II trial of ixabepilone administered daily for five days in children and young adults with refractory solid tumors: a report from the children's oncology group. Clin Cancer Res. 2010;16:750-4. [35] Widemann BC, Goodspeed W, Goodwin A, Fojo T, Balis FM, Fox E. Phase I trial and pharmacokinetic study of ixabepilone administered daily for 5 days in children and adolescents with refractory solid tumors. Journal of Clinical Oncology. 2009;27:550-6. [36] Blaney SM, Bernstein M, Neville K, Ginsberg J, Kitchen B, Horton T, et al. Phase I study of the proteasome inhibitor bortezomib in pediatric patients with refractory solid tumors: a Children's 33

Oncology Group study (ADVL0015). Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2004;22:4804-9. [37] Horton TM, Perentesis JP, Gamis AS, Alonzo TA, Gerbing RB, Ballard J, et al. A Phase 2 study of bortezomib combined with either idarubicin/cytarabine or cytarabine/etoposide in children with relapsed, refractory or secondary acute myeloid leukemia: a report from the Children's Oncology Group. Pediatric blood & cancer. 2014;61:1754-60. [38] Muscal JA, Thompson PA, Horton TM, Ingle AM, Ahern CH, McGovern RM, et al. A phase I trial of vorinostat and bortezomib in children with refractory or recurrent solid tumors: a Children's Oncology Group phase I consortium study (ADVL0916). Pediatric blood & cancer. 2013;60:390-5. [39] Lazzerini M, Martelossi S, Magazzu G, Pellegrino S, Lucanto MC, Barabino A, et al. Effect of thalidomide on clinical remission in children and adolescents with refractory Crohn disease: a randomized clinical trial. Jama. 2013;310:2164-73. [40] Lazzerini M, Martelossi S, Magazzu G, Pellegrino S, Lucanto MC, Barabino A, et al. Effect of thalidomide on clinical remission in children and adolescents with ulcerative colitis refractory to other immunosuppressives: Pilot randomized clinical trial. Inflammatory Bowel Diseases. 2015;21:173949. [41] Priolo T, Lamba LD, Giribaldi G, De Negri E, Grosso P, De Grandis E, et al. Childhood thalidomide neuropathy: a clinical and neurophysiologic study. Pediatric neurology. 2008;38:196-9. [42] Fleming FJ, Vytopil M, Chaitow J, Jones HR, Jr., Darras BT, Ryan MM. Thalidomide neuropathy in childhood. Neuromuscul Disord. 2005;15:172-6. [43] Jain P, Gulati S, Seth R, Bakhshi S, Toteja GS, Pandey RM. Vincristine-induced neuropathy in childhood ALL (Acute Lymphoblastic Leukemia) survivors: Prevalence and electrophysiological characteristics. Journal of child neurology. 2014;29:932-7. [44] Ramchandren S, Leonard M, Mody RJ, Donohue JE, Moyer J, Hutchinson R, et al. Peripheral neuropathy in survivors of childhood acute lymphoblastic leukemia. Journal of the peripheral nervous system : JPNS. 2009;14:184-9. [45] Harila-Saari AH, Vainionpaa LK, Kovala TT, Tolonen EU, Lanning BM. Nerve lesions after therapy for childhood acute lymphoblastic leukemia. Cancer. 1998;82:200-7. 34

[46] Lehtinen SS, Huuskonen UE, Harila-Saari AH, Tolonen U, Vainionpaa LK, Lanning BM. Motor nervous system impairment persists in long-term survivors of childhood acute lymphoblastic leukemia. Cancer. 2002;94:2466-73. [47] Earl HM, Connolly S, Latoufis C, Eagle K, Ash CM, Fowler C, et al. Long-term neurotoxicity of chemotherapy in adolescents and young adults treated for bone and soft tissue sarcomas. Sarcoma. 1998;2:97-105. [48] Ness KK, Hudson MM, Pui CH, Green DM, Krull KR, Huang TT, et al. Neuromuscular impairments in adult survivors of childhood acute lymphoblastic leukemia: associations with physical performance and chemotherapy doses. Cancer. 2012;118:828-38. [49] Ness KK, Jones KE, Smith WA, Spunt SL, Wilson CL, Armstrong GT, et al. Chemotherapyrelated neuropathic symptoms and functional impairment in adult survivors of extracranial solid tumors of childhood: results from the St. Jude Lifetime Cohort Study. Archives of physical medicine and rehabilitation. 2013;94:1451-7. [50] Hoffman MC, Mulrooney DA, Steinberger J, Lee J, Baker KS, Ness KK. Deficits in physical function among young childhood cancer survivors. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2013;31:2799-805. [51] Hartman A, van den Bos C, Stijnen T, Pieters R. Decrease in peripheral muscle strength and ankle dorsiflexion as long-term side effects of treatment for childhood cancer. Pediatric blood & cancer. 2008;50:833-7. [52] Reinders-Messelink HA, Schoemaker MM, Hofte M, Goeken LN, Kingma A, van den Briel MM, et al. Fine motor and handwriting problems after treatment for childhood acute lymphoblastic leukemia. Medical and pediatric oncology. 1996;27:551-5. [53] Wright MJ, Galea V, Barr RD. Proficiency of balance in children and youth who have had acute lymphoblastic leukemia. Physical therapy. 2005;85:782-90. [54] NCI. Common terminology criteria for adverse events v4.03. 2010. [55] Gilchrist LS, Marais L, Tanner L. Comparison of two chemotherapy-induced peripheral neuropathy measurement approaches in children. Supportive care in cancer : official journal of the Multinational Association of Supportive Care in Cancer. 2014;22:359-66. 35

[56] Gilchrist LS, Tanner L. The pediatric-modified total neuropathy score: a reliable and valid measure of chemotherapy-induced peripheral neuropathy in children with non-CNS cancers. Supportive care in cancer : official journal of the Multinational Association of Supportive Care in Cancer. 2013;21:847-56. [57] Lavoie Smith EM, Li L, Hutchinson RJ, Ho R, Burnette WB, Wells E, et al. Measuring vincristine-induced peripheral neuropathy in children with acute lymphoblastic leukemia. Cancer nursing. 2013;36:E49-60. [58] Bradfield SM, Sandler E, Geller T, Tamura RN, Krischer JP. Glutamic acid not beneficial for the prevention of vincristine neurotoxicity in children with cancer. Pediatric blood & cancer. 2015;62:1004-10. [59] Diouf B, Crews KR, Lew G, Pei D, Cheng C, Bao J, et al. Association of an inherited genetic variant with vincristine-related peripheral neuropathy in children with acute lymphoblastic leukemia. Jama. 2015;313:815-23. [60] Reinders-Messelink HA, Schoemaker MM, Snijders TA, Goeken LN, Bokkerink JP, Kamps WA. Analysis of handwriting of children during treatment for acute lymphoblastic leukemia. Medical & Pediatric Oncology. 2001;37:393-9. [61] Gilheeney SW, Lyden DC, Sgouros S, Antunes N, Gerald W, Kramer K, et al. A phase II trial of thalidomide and cyclophosphamide in patients with recurrent or refractory pediatric malignancies. Pediatric Blood and Cancer. 2007;49:261-5. [62] Moriyama B, Henning SA, Leung J, Falade-Nwulia O, Jarosinski P, Penzak SR, et al. Adverse interactions between antifungal azoles and vincristine: review and analysis of cases. Mycoses. 2012;55:290-7. [63] Teusink AC, Ragucci D, Shatat IF, Kalpatthi R. Potentiation of vincristine toxicity with concomitant fluconazole prophylaxis in children with acute lymphoblastic leukemia. Pediatric hematology and oncology. 2012;29:62-7. [64] Weintraub M, Adde MA, Venzon DJ, Shad AT, Horak ID, Neely JE, et al. Severe atypical neuropathy associated with administration of hematopoietic colony-stimulating factors and vincristine. Journal of Clinical Oncology. 1996;14:935-40. 36

[65] Bhushan B, Bhargava A, Kasundra GM, Shubhakaran K, Sood I. Guillain-Barre syndrome in acute lymphoblastic leukemia: Causal or coincidental. Journal of pediatric neurosciences. 2015;10:646. [66] Brigo F, Balter R, Marradi P, Ferlisi M, Zaccaron A, Fiaschi A, et al. Vincristine-related neuropathy versus acute inflammatory demyelinating polyradiculoneuropathy in children with acute lymphoblastic leukemia. Journal of child neurology. 2012;27:867-74. [67] Ceppi F, Langlois-Pelletier C, Gagne V, Rousseau J, Ciolino C, De Lorenzo S, et al. Polymorphisms of the vincristine pathway and response to treatment in children with childhood acute lymphoblastic leukemia. Pharmacogenomics. 2014;15:1105-16. [68] Trobaugh-Lotrario AD, Smith AA, Odom LF. Vincristine neurotoxicity in the presence of hereditary neuropathy. Medical and pediatric oncology. 2003;40:39-43. [69] Mokhtar GM, Shaaban SY, Elbarbary NS, Fayed WA. A trial to assess the efficacy of glutamic acid in prevention of vincristine-induced neurotoxicity in pediatric malignancies: A pilot study. Journal of pediatric hematology/oncology. 2010;32:594-600. [70] Akbayram S, Akgun C, Dogan M, Sayin R, Caksen H, Oner AF. Use of pyridoxine and pyridostigmine in children with vincristine-induced neuropathy. Indian J Pediatr. 2010;77:681-3. [71] Cavaletti G, Cornblath DR, Merkies IS, Postma TJ, Rossi E, Frigeni B, et al. The chemotherapyinduced peripheral neuropathy outcome measures standardization study: from consensus to the first validity and reliability findings. Annals of oncology : official journal of the European Society for Medical Oncology / ESMO. 2013;24:454-62. [72] Postma TJ, Heimans JJ, Muller MJ, Ossenkoppele GJ, Vermorken JB, Aaronson NK. Pitfalls in grading severity of chemotherapy-induced peripheral neuropathy. Annals of oncology : official journal of the European Society for Medical Oncology / ESMO. 1998;9:739-44. [73] Klepper SE. Measures of pediatric function: Child Health Assessment Questionnaire (C-HAQ), Juvenile Arthritis Functional Assessment Scale (JAFAS), Pediatric Outcomes Data Collection Instrument (PODCI), and Activities Scale for Kids (ASK). Arthritis care & research. 2011;63 Suppl 11:S371-82.

37

[74] Cavaletti G, Frigeni B, Lanzani F, Mattavelli L, Susani E, Alberti P, et al. ChemotherapyInduced Peripheral Neurotoxicity assessment: a critical revision of the currently available tools. European journal of cancer. 2010;46:479-94. [75] Beutler AS, Kulkarni AA, Kanwar R, Klein CJ, Therneau TM, Qin R, et al. Sequencing of Charcot-Marie-Tooth disease genes in a toxic polyneuropathy. Annals of neurology. 2014;76:727-37. [76] Dobrozsi S, Flood VH, Panepinto J, Scott JP, Brandow A. Vitamin B12 deficiency: the great masquerader. Pediatric blood & cancer. 2014;61:753-5. [77] Hershman DL, Lacchetti C, Dworkin RH, Lavoie Smith EM, Bleeker J, Cavaletti G, et al. Prevention and management of chemotherapy-induced peripheral neuropathy in survivors of adult cancers: American Society of Clinical Oncology clinical practice guideline. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2014;32:1941-67. [78] Sman AD, Hackett D, Fiatarone Singh M, Fornusek C, Menezes MP, Burns J. Systematic review of exercise for Charcot-Marie-Tooth disease. Journal of the peripheral nervous system : JPNS. 2015;20:347-62.

38

Funding This work was supported by the National Health and Medical Research Council (APP1114801), the Royal Australasian College of Physicians and the Cancer Institute of the New South Wales (grant number 14/TPG/1-05).

39

40

41

42

Highlights •

CIPN is common and can be long lasting

•

Characteristics and natural history of neuropathy are specific to chemotherapy agent

•

Significant variation in incidence may be due to challenges in clinical assessment

•

Risk factors for CIPN may be disease, treatment or patient related

•

Standardised age-appropriate assessment protocols are essential for future research

43