Prospective assessment of vincristine-induced peripheral neuropathy in paediatric acute lymphoblastic leukemia

(cid:1) Vincristine treatment produces a sensorimotor neuropathy in children, which develops early in the treatment course. (cid:1) Both neurophysiological and clinical studies provide evidence for motor predominance. (cid:1) Baseline nerve proﬁles are important to identify early nerve changes and allow neuroprotection strategies.


Introduction
Recent decades have seen an increase in the incidence of childhood cancer diagnoses (Howlader et al., 2020).However concurrently, there has also been reduction in mortality and improvement in 5-year survival rates for most paediatric cancers https://doi.org/10.1016/j.clinph.2023.08.002 1388-2457/Ó 2023 International Federation of Clinical Neurophysiology.Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).(Howlader et al., 2020).As a result, the population of survivors of childhood cancer is steadily growing.Acute lymphoblastic leukemia (ALL) is the most common childhood malignancy with a 5year survival of > 90% (Howlader et al., 2020).In-light of this prognosis, it is critical to consider and characterise long-term side effects of ALL treatment.
Vincristine is a vinca alkaloid that is a mainstay chemotherapy treatment for ALL (Gilchrist 2012).Vincristine-induced peripheral neuropathy (VIPN) is a very common adverse effect, presenting in up to 78%-100% of patients exposed to vincristine treatment (Kandula et al. 2016).VIPN typically manifests with motor impairments including foot drop, ataxia and gait abnormality, although sensory symptoms including paraesthesia may also be present (Kandula et al. 2016).
Despite this pervasive and significant side effect, very little is still known about the natural history of VIPN development, progression and reversibility.One of the main reasons may be due to a lack of standardised paediatric neuropathy assessment tools.Clinical detection of VIPN in children has been limited by their inability to articulate symptoms.The current gold standard for objective neurophysiologic assessment of neuropathy is nerve conduction studies (Kandula et al. 2017).However, their lack of sensitivity suggests that changes may not appear until later in the neuropathy progression when the damage may be irreversible (Kokotis et al. 2016).Traditionally, in paediatric VIPN, neurophysiology studies are done following the onset of symptoms and neuropathy is defined relative to broad age-matched reference ranges.The lack of baseline studies also reduces sensitivity to detect early neuropathy.Nerve excitability studies provide more comprehensive assessment of nerve function than traditional nerve conduction studies by providing information on ion channel properties and resting membrane potential (Kiernan et al. 2020).Moreover, excitability studies have demonstrated change in nerve properties prior to the onset of neuropathy in patients receiving other neurotoxic chemotherapies (Park et al. 2009;Park et al. 2011).
In order to address the gap in knowledge of paediatric VIPN disease progression, pathophysiology and reversibility, this longitudinal study from baseline, through to the follow-up period after completion of chemotherapy, aimed to investigate, detect early neuropathy and characterise the natural history of VIPN throughout its disease course using clinical and neurophysiological assessment measures.

Study design
Paediatric patients (<18 years old) diagnosed with ALL who were about to commence vincristine-based treatment at the Syd-ney Children's Hospital, Australia were recruited to the study from May 2015 to November 2016.The study was approved by the Sydney Children's Hospitals Network Human Research Ethics Committee.Written informed consent was obtained from each participant or their parent/guardian in accordance with the Declaration of Helsinki.Patient demographic information were collected from medical records.
All patients received vincristine induction as part of the Associazone Italiana Ematologia Oncologia Pediatrica Berlin-Frankfurt-Münster (AIEOP-BFM Study 9) protocol, with reinduction depending on ALL risk level (European Clinical Trials Database 2007-004270-43).All patients received vincristine (1.5 mg/m 2 , capped at 2 mg) administered during the 35-day induction on days 8, 15, 22, 29.Standard risk ALL patients subsequently received a M Phase (no additional neurotoxic treatment) followed by a 35-day reinduction with vincristine (1.5 mg/m 2 , capped at 2 mg) administered on days 8, 15, 22, 29.Patients were assessed at baseline (patients who had received a single vincristine dose were also included), post-induction, prior to commencing reinduction, postreinduction and at follow-up.The treatment protocol and assessment timepoints are illustrated in Fig. 1.

Neuropathy assessment
Patients were clinically assessed by a paediatric neurologist (TK) at each timepoint.A motor and sensory neuropathy score was graded using the Balis Pediatric Scale of Peripheral Neuropathy (Smith et al. 2020) based on clinical impression as well as patient or parent report of symptoms.This validated instrument assesses neurotoxicity in children and grades symptoms on a scale of 0-4, with higher grades indicating worse neuropathy (Smith et al. 2020).Patients were categorized using summed sensory and motor Balis grades into no/mild VIPN (sum score 0-2) or significant VIPN (sum score 3-8) groups.
Parents of participants completed the Pediatric Quality of Life Inventory, parent report (PedsQL) at each assessment timepoint.The PedsQL is a validated generic health related quality of life (QoL) measure consisting of 4 core scales: physical function, emotional function, social function and school function (Varni et al. 2003).Overall scores are averaged across cores and range from 0-100 with higher scores indicating higher QoL.
Motor and sensory nerve excitability studies were conducted using threshold tracking software (QtracÓ Institute of Neurology, Queen Square, UK) according to consensus guidelines and protocols described previously (Kiernan et al. 2020;Kiernan et al. 2000).The stimulating cathode was positioned over the median nerve approximately at the proximal wrist crease, the anode positioned 5-10 cm proximal on the lateral aspect of the forearm and an earthing electrode was placed on the dorsum of the hand  Nerve excitability studies were carried out under general anaesthesia at the same time as participants received other medical procedures.For participants who did not require general anaesthesia for their medical procedures (n = 4), studies were undertaken whilst conscious with distraction.
Standard motor and sensory excitability TROND protocols were conducted as previously described (Kiernan et al. 2000;Kiernan et al. 2001), with a sequence of parameters recorded including a stimulus response curve, strength-duration time constant (SDTC), threshold electrotonus (TE), current-threshold relationship (IV) and recovery cycle (RC).
The stimulus-response curve was generated by increasing the stimulus intensity in a stepwise manner from zero in 2% increments until a maximal response amplitude was generated.The current required to produce a defined target amplitude (set to 40% of maximal amplitude) was tracked following different stimulus patterns, including: changes in stimulus width (strengthduration relationship), 100 ms subthreshold polarising currents set to ± 40% of control threshold (threshold electrotonus), 200 ms polarising currents, with current strength stepped in 10% intervals from 50% to À100% of control threshold (current threshold relationship) and sequential alteration of the interval between a supramaximal and conditioning stimulus from 2 to 200 ms (recovery cycle).
Key parameters were derived as in Kiernan et al. (Kiernan et al. 2000;Kiernan et al. 2001).Strength duration time constant (SDTC) was defined as the x-intercept of the linear relationship between the stimulus intensity and stimulus width according to Weiss' law (Weiss 1901;Bostock 1983).Threshold electrotonus parameters were assessed at intervals during (between 10-100 ms) and after the application of depolarizing (TEd) or hyperpolarizing (TEh) current.TEd peak reflected the mean maximal percentage threshold reduction in response to depolarising current averaged over 20 ms, and S2 accommodation was the difference between TEd peak and TEd at 90-100 ms, reflecting accommodation towards baseline threshold.Recovery cycle parameters included superexcitability as the minimum mean threshold change of three adjacent points and subexcitability as the minimum mean threshold change after interstimulus intervals of 10 ms.

Statistical analysis
All results were expressed as mean ± standard deviation.Differences between timepoints were analysed within individuals using paired analyses.Unless otherwise specified, statistical significance was defined as a P value of < 0.05.Significance for nerve excitability results were corrected for multiple comparisons (significance threshold P < 0.0125).All statistical analyses were performed using Stata version 14 (StataCorp, College Station, Texas, USA).

Results
Thirty-one paediatric patients diagnosed with ALL who were due to commence vincristine treatment were recruited to the study (Table 1).Patients had a mean age of 6.8 ± 4.4 years and were predominantly female (61.3%, n = 19).Mean baseline QoL scores as assessed using the PedsQL parent report was 83.6 ± 12.3 out of 100.
Majority of patients presented at baseline with no motor (83.9%, n = 26) or sensory (83.9%, n = 26) neuropathy symptoms as assessed with the Balis motor and sensory neuropathy scales.Average age of patients with baseline Balis scores > 0 was 4.0 ± 2.1 years.At baseline, sensory deficits were pain related and motor deficits were mobility related.There were no differences in age, baseline QOL, baseline motor Balis or baseline sensory Balis scores between patients who were assessed prior to vincristine initiation, and after 1 vincristine dose (all P > 0.05).Baseline nerve excitability parameters were not significantly different (with the exception of sensory latency, P < 0.0125) between patients who had their studies completed awake or under anaesthesia.

Clinical characteristics
There was significant development of motor VIPN from T0 to T1 (mean Balis motor score 0.3 ± 0.8 vs 2.1 ± 0.9, P < 0.001).Difficulty with walking, crawling or standing up unassisted was reported in patients with motor Balis grade ! 1.By T2, motor symptoms had significantly improved and were comparable to T0 levels (P > 0.05).Some patients still reported minor functional difficulties including occasional stumbles or difficulty with jogging, however only 36.0%(n = 9) of patients still had a motor Balis score > 0 at T2, compared to 92.0% (n = 23) at T1 (Fig. 2).In contrast, there was no significant increase in sensory VIPN from T0 to T1 (mean Balis sensory score 0.2 ± 0.5 vs 0.6 ± 1.0, P > 0.05), although 36.0%(n = 9) of patients had a sensory Balis score > 0 at T1, compared to 20% at T0 (n = 5) and 16% at T2 (n = 4).Patients with sensory Balis grade ! 1 reported pain in the legs, numbness in the extremities or sensitivity to temperatures or textures.Overall, vincristine induction was associated with significantly more prominent motor VIPN symptoms than sensory symptoms with 92.0% (n = 23) of patients having a motor Balis score > 0 compared to 36.0%(n = 9) having sensory Balis score > 0 at T1 (v 2 = 17.0,P < 0.01).

Neurophysiological characteristics
In line with the development of symptomatic motor VIPN by T1 in the majority of patients, there was a significant decline in motor nerve peak CMAP amplitude from T0 to T1 (diff = 2.1 ± 1.7 mV, P < 0.001).This reduced peak amplitude persisted at the T2 timepoint and remained significantly reduced compared to baseline amplitude (diff = 1.6 ± 2.6 mV, P < 0.01) (Fig. 3a).Vincristine induction also resulted in extensive changes in motor nerve excitability properties including in recovery cycle (RC) and depolarising threshold electrotonus (TE) parameters (Table 2).Overall, there was increased threshold change in depolarizing TE and increased superexcitability at T1, which persisted at T2 (Fig. 4, Table 2).
Subgroup analysis was conducted between patients who had developed none/mild VIPN (n = 12) compared to significant VIPN (n = 14) during vincristine induction.There were no significant differences in the extent of change in motor or sensory nerve excitability parameters between patients who had none/mild and significant VIPN (Supplementary Tables S1 and S2).

Characteristics of VIPN from reinduction
In order to examine the development of VIPN from vincristine reinduction, twenty-two patients who had completed pre (T2) and post-reinduction (T3) assessments were evaluated.T3 assessment was 1.8 ± 1.1 months following T2 assessment.

Neurophysiological studies
Sensory and motor nerve excitability studies were also undertaken at T2 and T3 to assess changes in neurophysiological profile from vincristine reinduction.There was no significant reduction in median CMAP or SNAP amplitudes at T3 compared to T2 (both P > 0.05).Changes in RC and TE parameters in the motor excitability studies were minimal (Table 5).Sensory nerves demonstrated increased latency, accommodation in TE and increased superexcitability at T3 compared to T2 (Table 5).Results presented as mean ± SD.T1 was completed 30.0 ± 9.1 days after T0, T2 was completed 3.3 ± 1.3 months after T1. * Denotes significant difference compared to T0 following correction for multiple comparison (P < 0.0125).
+ Denotes significant difference compared to T1 following correction for multiple comparison (P < 0.0125).
+ Denotes significant difference compared to T1 following correction for multiple comparison (P < 0.0125).
Nerve excitability studies were completed in a subset of patients at follow-up (n = 19).There was evidence of persistent changes compared to T0 in both motor and sensory axons (Table 6).On motor studies, there was evidence of sustained increase in hyperpolarising TE and accommodation, as well as reduced superexcitability (Table 6).Sensory excitability studies demonstrated a sustained reduction in median SNAP (59.8 ± 22.4 lV vs 42.8 ± 15.3 lV, P = 0.01), as well as increased latency (2.6 ± 0.4 vs 2.8 ± 0.4, P < 0.001) but no changes in TE or RC excitability parameters at T4 compared to T0 (Table 6).
Results presented as mean ± SD.T1 was completed 30.0 ± 9.1 days after T0, T2 was completed 3.3 ± 1.3 months after T1. n.s denotes non-significance following correction for multiple comparison (P < 0.0125).were no significant difference in parameters between patients with and without residual VIPN at T4.There were also no significant differences in the extent of change in key motor or sensory nerve excitability variables from T0 to T4 between patients with or without VIPN at follow-up.Neither baseline (T0) or post-induction (T1) Balis grades, nor the extent of early changes in motor and sensory nerve excitability variables (T0 to T1) were able to predict presence of persistent VIPN symptoms at follow-up (all P > 0.05).

Discussion
This study comprehensively details clinical and neurophysiological changes in nerve function in children undergoing vincristine treatment.Overall, there was evidence of both sensory and motor VIPN development, with clinical symptoms and nerve excitability profiles changing in tandem.However, motor neuropathy development was more prominent than sensory, and this was reflected in both clinical and neurophysiological assessments.There was evidence of greater VIPN development during vincristine induction with less although still significant VIPN development during reinduction.In total, this study highlights the early impact of vincristine-containing treatment protocols on the functional properties of both sensory and motor nerves and suggests that follow-up may be required to ensure that nerve function recovers in childhood cancer survivors.In addition, the study provides a timecourse for the development of neuropathic complications of vincristine, that may be utilised to promote neuroprotective strategies.
Results Presented as mean ± SD.T4 was completed 7.8 ± 5.5 months after T0. * Denotes significant difference compared to T0 following correction for multiple comparison (P < 0.0125).
toms by end of vincristine treatment (Rodwin et al. 2022;Yildiz and Temucin, 2016).These findings are broadly similar to the prevalence of motor (92%) and sensory symptoms (36%) at end of induction in the present study.Post-treatment, the prevalence of persisting neuropathy varies depending of assessment measure from 12% when assessing using the pediatric-modified total neuropathy score (ped-mTNS) (Gilchrist et al. 2017), to 68% when assessing using other functional evaluations (including monofilament and dynamometers) (Rodwin et al. 2022).The present study assessed VIPN using the Balis scale (graded by neurologist via clinical impression) and found 48% of patients had persisting VIPN at 8 months post treatment.
Prior studies investigating electrophysiologic properties of paediatric VIPN have also demonstrated predominantly motor nerve involvement (Kavcic et al. 2017;Yildiz and Temucin, 2016;Jain et al. 2014;Jeong et al. 2023), with one study finding 92% of patients had motor and 8% had sensory nerve abnormalities compared to controls, within 2 months of vincristine treatment (Yildiz and Temucin, 2016).Persisting VIPN has also been demonstrated in electrophysiological studies, with a study finding 34% of patients demonstrated motor nerve abnormalities up to 2 years post vincristine treatment, with no patients demonstrating sensory nerve abnormalities (Jain et al. 2014).The present study showed a significant decline in median nerve CMAP early during vincristine treatment, but recovery in motor amplitudes by 8 months post treatment.The difference in electrophysiologic evidence of VIPN at follow-up may be due the difference the duration of follow-up and the in nerves assessed, with studies suggesting motor amplitude decline may be more persistent in lower-limb compared to upper-limb motor nerves (Jeong et al. 2023;Jain et al. 2014).

Phenotypic differences between adult and paediatric VIPN
Unlike VIPN manifestation in children, sensory neuropathy is the most common presentation of VIPN in adults (Caccia et al. 1977;Pal 1999;DeAngelis et al. 1991;Casey et al. 1973), affecting 75-78% of patients (Pal 1999;Haim et al. 1994), whereas motor neuropathy is reported in 19-45% of patients (Haim et al. 1994;Pal 1999).Electrophysiological studies of VIPN in adults have been limited in recent years, however early studies demonstrated motor nerve involvement, although sensory nerves were still predominantly affected (Casey et al. 1973;McLeod and Penny, 1969;Bradley et al. 1970;Guiheneuc et al. 1980;DeAngelis et al. 1991).This is contrary to the paediatric cohort in the present study where motor VIPN occurred in the majority of patients (92%) compared to sensory VIPN in 35%.Previous studies have also highlighted the motor predominance of paediatric VIPN, compared to the expression of symptoms in adults (Rodwin et al. 2022;Yildiz and Temucin, 2016;Courtemanche et al. 2015).
There are several potential reasons for the differences between the sensory-motor presentation of VIPN in children and adults.The biophysical properties of motor and sensory nerves in adult and paediatric cohorts are different (Kandula et al. 2020b;Farrar et al. 2013), and this may be reflected in the different toxicity effects of vincristine in these cohorts.A key finding in this study is the acute onset of clinical motor symptoms and 30% loss of CMAP within one month of vincristine treatment, a pattern of nerve damage that has not been seen in adults.This may suggest that paediatric motor nerves have increased susceptibility to damage.However, studies in children treated with cisplatin, a traditionally sensory-neurotoxic chemotherapy in adults, also demonstrated sensory nerve changes in children (Kandula et al., 2020a), highlighting that motor predominance does not occur with all agents.It is likely that the interaction between toxic agents and maturing nerves are complex, and future studies will be needed to probe mechanisms.
Further, a possible consideration is the comparative visibility of motor VIPN symptoms compared to sensory symptoms, which may partially account for the higher rates of motor VIPN, particularly in younger age groups with proxy reporting.Complexities of assessing neuropathy in children are further compounded with the complexities of other symptoms in ALL patients (Hunger and Mullighan, 2015).For example, a clinical presentation of a patient 'going off their feet' may be due to steroid-induced myopathy, neuropathy or systemic ill health.However, changes in neurophysiological studies found in this study provide evidence of vincristine's toxic effects on nerve function, particularly in motor nerves.

Nerve excitability studies in vincristine treated patients
While prior studies have investigated nerve excitability parameters in neurotoxic chemotherapy-treated patients, changes in sensory and motor axonal properties in vincristine-treated children during treatment have not been previously assessed.This study has demonstrated changes in motor, and to a lesser extent sensory nerve parameters following vincristine treatment.Increased threshold change in depolarizing TE and increased superexcitability were seen in motor axons, with changes in sensory parameters mostly found in the recovery cycle.Prior studies have suggested vincristine motor neuropathy is associated with prolonged distal latency (Kandula et al. 2017;Bradley et al. 1970) potentially due to axonal transport dysfunction (Park et al. 2008), and this was reflected in the present study where latency was significantly prolonged during induction, but not at follow-up.Similarly, experimental studies of motor axon excitability properties following nerve injury demonstrated rapid decline in CMAP with latency prolongation (Moldovan et al. 2008).Further, subsequent widespread excitability changes occurred in both depolarizing and hyperpolarizing TE with enhanced superexcitability in the recovery cycle (Moldovan et al. 2008).These patterns are similar to the excitability changes demonstrated in motor nerves following vincristine, and potentially suggest that they represent functional changes associated with axonal degeneration.
Neurophysiological studies are only able to measure the properties of conducting axons.Consequently, reduction in motor nerve amplitudes coupled with extensive excitability parameter change suggests early reduction in the number of conducting motor axons as well as functional changes in remaining axons.In sensory axons there was a different timecourse of change with a reduction of sensory amplitude seen following vincristine induction, with less functional change in remaining axons.Motor and sensory axons have different biophysical properties and differential response to toxicity (Kandula et al., 2020a;Lin et al. 2002).Further, in excitability studies of adult patients treated with paclitaxel, a compound known to cause significant sensory neuropathy through damage to microtubules, sensory amplitude decreased without changes in axonal function (Park et al. 2011), suggesting that nerve excitability techniques were not able to ascertain any functional changes prior to axonal loss.
In terms of suggesting specific pathophysiological mechanisms of axonal damage following vincristine, while broadly the pattern of increased threshold electrotonus and increased superexcitability in motor axons is similar to some effects of decreased potassium currents (Tomlinson et al. 2010;Tomlinson et al. 2012), these changes are more substantial than the vincristine induced changes.Further, although it has been demonstrated that vincristine affects axonal excitability in experimental studies (Ravula et al. 2007;Schappacher et al. 2019), there is no experimental evidence of a direct effect of vincristine on potassium channel activity.In addition, the pattern of changes in motor axons following the induction course is also similar to the effects of reduced serum potassium levels in normal axons (Boërio et al. 2014).Largescale studies of pediatric ALL patients suggest that significant hypokalemia occurrs in 14% of ALL patients during induction therapy (Miller et al. 2022), suggesting that more subtle alteration in potassium levels may occur in a substantial fraction of ALL patients on vincristine therapy.Taken in total, excitability alterations in vivo likely reflect multiple phenomena, as vincristine produces different effects on axonal excitability depending on the site and timecourse of administration in experimental models (Ravula et al. 2007, Schappacher et al. 2019).

Recoverability of VIPN
Consistent with prior studies (Courtemanche et al. 2015), follow-up analyses demonstrate evidence of persisting VIPN symptoms, with motor symptoms again more prevalent than sensory.Further, while there was evidence of partial normalisation in both motor and sensory nerves at follow-up compared to baseline, there was persistent excitability change in motor axons and persistently reduced sensory amplitudes.In contrast, prior cross-sectional studies have suggested that paediatric cancer patients treated with vincristine do not have widespread excitability changes at longterm follow-up (Kandula et al., 2020a).However, it is likely that these subtle but persisting changes are more apparent in a longitudinal within-patient study design.Similarly, long-term follow-up studies investigating VIPN at 8.5 years post treatment (Kandula et al. 2018) found more evidence of persisting sensory VIPN rather than motor VIPN, as well as persisting reduction in sensory sural nerve amplitude (Kandula et al. 2018).This suggests that VIPN may take years to recover, and despite motor nerves being more acutely affected by vincristine treatment, sensory nerves may demonstrate less recovery.There are no effective pharmacologic strategies available to protect nerve function in children with VIPN (Anghelescu et al. 2011;Bradfield et al. 2015).Exercise programs, particularly during treatment may preserve motor and physical activity ability (Gaser et al. 2022), however further research needs to investigate the effect and viability on younger cohorts.
Furthermore, while the present study did not find significant associations between residual VIPN and QoL at 8-months post vincristine treatment, previous studies have demonstrated children with VIPN have lower QoL scores than those without (van de Velde et al. 2021).However, this may be a short-term effect, with this decline seen earlier in the treatment course (van de Velde et al. 2021), and not 8-months post vincristine treatment in the present study.Nonetheless, long-term outcomes in vincristine-treated children need to be continually monitored, with long-term VIPN demonstrated to be associated with reduced fine motor (Sabarre et al. 2014) and gait function (Gilchrist and Tanner, 2016), and this can have further detrimental consequences on long-term QoL.

VIPN outcome measures
The lack of correlation between Balis scores and nerve excitability profiles may further highlight the need to develop more sensitive VIPN assessment measures in the paediatric cohort (Smith et al. 2020) in order to monitor symptom development.Prior studies of paediatric VIPN have also highlighted discrepancies between electrophysiological evidence of VIPN and clinical symptoms (Jain et al. 2014;Tunjungsari et al. 2021).Furthermore, even in adult cohorts, discordance has been demonstrated between patient reported and neurophysiological assessment of neuropathy (Timmins et al. 2020;Alberti et al. 2014) likely due to the inability of any single measure to capture the full spectrum of neuropathy manifestations.The utility and translation of these outcomes are distinct, with neurophysiological studies used to provide a specific measure of nerve involvement, and clinical grades to establish symptom severity and impact on function.Therefore, the implementation of both clinical and neurophysiological measures in research settings are recommended in order to capture the full spectrum of VIPN.
Other outcome measures to assess paediatric VIPN have also been previously evaluated.The ped-mTNS, a composite measure consisting of 8 neuropathy signs and symptoms, has demonstrated validity and reliability in paediatric cancer cohorts (Gilchrist et al. 2014;Gilchrist and Tanner, 2013;Gilchrist et al. 2009;Gilchrist and Tanner, 2018).However, these validation studies were completed on children aged !5 years, due to the difficulties of neurological assessment in younger children.In the present study, 45% of patients were under 5 years old, meaning that other assessment tools would be required to capture the burden of VIPN.The P-CIN, a paediatric patient-reported outcome measure has been validated for assessing neuropathy in paediatric cancer patients (Smith et al. 2021), although only in patients aged !6 years.Pain scales including the Pediatric Neuropathic Pain ScaleÓ (Lavoie Smith et al. 2015) and FACESÓ pain scale (Lavoie Smith et al. 2013) have been used to investigate vincristine-associated pain.However, these studies were also completed on children aged !5 years, and identified pain as an uncommon and mild feature of paediatric VIPN (Lavoie Smith et al. 2015).In total, there is a lack of suitable peripheral neuropathy outcome measures available for the paediatric cancer cohort, particular in younger children.Robust tools to detect and monitor neuropathy development is essential to identify and minimise long-term nerve damage.

Limitations
While attrition at each timepoint may hinder the ability to detect changes, investigators for the present cohort endeavoured to capture all timepoints by completing investigations at participants' convenience, although due to the ill-health and burden of other hospital admissions and procedures associated with cancer treatment, some timepoints were missed.Separately, the electrophysiological investigations were completed in the hands, as this was a surrogate marker of generalised exposure of peripheral nerves to neurotoxic therapy.Excitability protocols have been more robustly developed in the hands, with limited studies investigating lower-limb nerves, due to floor effects.Finally, at this stage, the results relating to persisting VIPN are limited to the follow-up period of 8 months, ongoing longer-term longitudinal data may provide a clearer profile of reversibility.
In summary, the present series of analyses illustrate the natural history of paediatric VIPN from both clinical and neurophysiological perspectives.This study highlights the need to conduct baseline neurological assessment for paediatric ALL patients commencing vincristine treatment.VIPN signs and symptoms develop early in the treatment course and baseline nerve profiles would allow for early detection of nerve changes and timely implementation of neuroprotection strategies.

Fig. 3 .
Fig. 3. Motor and sensory peak amplitude changes from vincristine induction.Error bars signify standard error of the mean.*Significant difference at P < 0.05.**Significant difference at P < 0.01.

Fig. 4 .
Fig. 4. Motor and Sensory nerve excitability changes from vincristine induction Error bars indicate standard error.

Table 1
Baseline patient demographic information.

Table 3
Sensory nerve excitability values from vincristine induction.

Table 4
Difference in motor nerve excitability parameters between age groups.

Table 5
Motor and sensory nerve excitability from vincristine reinduction.Motor and sensory nerve excitability studies at follow-up compared to baseline.
* Denotes significant difference compared following correction for multiple comparison (P < 0.0125).