Ò

PAIN 155 (2014) 2461–2470

www.elsevier.com/locate/pain

Comprehensive review

Incidence, prevalence, and predictors of chemotherapy-induced

peripheral neuropathy: A systematic review and meta-analysis
Marta Seretny a,⇑, Gillian L. Currie b, Emily S. Sena b, Sabrina Ramnarine a, Robin Grant c,
Malcolm R. MacLeod b, Leslie A. Colvin c, Marie Fallon a
a
Cancer Research UK, University of Edinburgh, Edinburgh, Scotland, UK
b
Centre for Clinical Brain Sciences, University of Edinburgh, Edinburgh, Scotland, UK
c
Western General Hospital, University of Edinburgh, Edinburgh, Scotland, UK

Sponsorships or competing interests that may be relevant to content are disclosed at the end of this article.

a r t i c l e i n f o a b s t r a c t

Article history: Chemotherapy-induced peripheral neuropathy (CIPN) is a disabling pain condition resulting from chemo-
Received 19 August 2014 therapy for cancer. Severe acute CIPN may require chemotherapy dose reduction or cessation. There is no
Received in revised form 12 September 2014 effective CIPN prevention strategy; treatment of established chronic CIPN is limited, and the prevalence
Accepted 16 September 2014
of CIPN is not known. Here we used a systematic review to identify studies reporting the prevalence of
CIPN. We searched Embase, Medline, CAB Abstracts, CINAHL, PubMed central, Cochrane Library, and
Web of Knowledge for relevant references and used random-effects meta-regression to estimate overall
Keywords:
prevalence. We assessed study quality using the CONSORT and STROBE guidelines, and we report findings
Chemotherapy-induced peripheral
neuropathy (CIPN)
according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidance.
Prevalence We provide a qualitative summary of factors reported to alter the risk of CIPN. We included 31 studies
Risk factors with data from 4179 patients in our analysis. CIPN prevalence was 68.1% (57.7–78.4) when measured
Systematic review in the first month after chemotherapy, 60.0% (36.4–81.6) at 3 months and 30.0% (6.4–53.5) at 6 months
Meta-analysis or more. Different chemotherapy drugs were associated with differences in CIPN prevalence, and there
was some evidence of publication bias. Genetic risk factors were reported in 4 studies. Clinical risk
factors, identified in 4 of 31 studies, included neuropathy at baseline, smoking, abnormal creatinine
clearance, and specific sensory changes during chemotherapy. Although CIPN prevalence decreases
with time, at 6 months 30% of patients continue to suffer from CIPN. Routine CIPN surveillance during
post-chemotherapy follow-up is needed. A number of genetic and clinical risk factors were identified that
require further study.
Ó 2014 The Authors. Published by Elsevier B.V. on behalf of International Association for the Study of
Pain. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/
by-nc-nd/3.0/).

1. Introduction CIPN is poorly understood, and treatments to prevent CIPN are

inadequate. Meta-analyses of clinical trials for CIPN prevention
Chemotherapy-induced peripheral neuropathy (CIPN) is a dis- report inconclusive results [1,49]. Treatment options for estab-
abling side effect of several commonly used antineoplastic agents. lished CIPN are also limited. Clinical trials of antiepileptic or anti-
The development of CIPN may require chemotherapy dose reduc- depressant agents to treat other neuropathic pain conditions
tion or cessation, which can increase cancer-related morbidity have generally been negative [30,41,54,55]. Only 1 recent, dou-
and mortality [17,31]. CIPN is a predominantly sensory neuropathy ble-blind, randomized controlled trial showed improvement in
that may be accompanied by motor and autonomic changes [62]. CIPN symptoms after 5 weeks of treatment with duloxetine [57].
Similar to other neuropathic pain conditions, pain in CIPN can be Understanding of the epidemiology of CIPN is also limited [37].
stimulus dependent or independent [66]. The pathophysiology of Previous studies have largely focussed on individual chemothera-
peutic agents, with reported CIPN incidence rates ranging from
19% to more than 85% [23]. Annually 165,544 patients survive can-
⇑ Corresponding author at: Cancer Research UK, Institute of Genetics and
cer in the United Kingdom, and more than 1 million in the United
Molecular Medicine, University of Edinburgh, Crew Road South, EH4 2XR
States [12,44]. It is therefore important to provide a more precise
Edinburgh, Scotland, UK. Tel.: +44 7724528933.
E-mail address: [email protected] (M. Seretny). measure of the prevalence of CIPN to allow appropriate resource

http://dx.doi.org/10.1016/j.pain.2014.09.020
0304-3959/Ó 2014 The Authors. Published by Elsevier B.V. on behalf of International Association for the Study of Pain.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).
 Ò
2462 M. Seretny et al. / PAIN 155 (2014) 2461–2470

allocation and research planning, and to inform patient decisions 2.3. Data extraction and quality assessment
about treatment. Understanding risk factors (including genetic risk
factors) for CIPN may guide future research and treatment. We extracted data to a bespoke form, recording the prevalence
Previous reviews of CIPN have combined narrative review with or incidence of CIPN, and any reported risk factors or predictors of
expert opinion, with potential risk of bias [15,28,29]. Here we pres- CIPN. We included all relevant outcomes determined after the end
ent what we believe to be the first systematic review and meta- of chemotherapy, noting the time (in relation to the end of chemo-
analysis of the incidence and prevalence of CIPN. We also aimed therapy) at which these were assessed. Where information was
to assess the influence of potential publication bias on our estima- incomplete we contacted authors by email. Two investigators
tion of CIPN measures, and to seek empirical evidence of the (M.S. and S.R.) extracted data, which were then entered into the
impact of study design factors. study database. Discrepancies were resolved by discussion and
agreement with a third reviewer (M.F.).
2. Methods We assessed study quality according to the PRISMA guidelines
[43]. We evaluated risk of bias in individual studies using the fol-
2.1. Search strategy lowing criteria: investigator blinding of any type, presence of a
control group, use of externally validated instruments for CIPN
We searched Embase, Medline, CAB Abstracts, CINAHL, assessment, clear description of statistical methods used to iden-
PubMed central, Cochrane Library and Web of Knowledge in July tify CIPN predictors, and description of longitudinal follow up.
2013 for English-language references. Searches were not limited Adherence of each study to relevant reporting criteria (STROBE or
by date restrictions. Search terms were free text and included; CONSORT) was assessed [2,61]. We assessed the risk of bias for
[‘‘Chemotherapy Induced Peripheral Neuropathy’’ OR ‘‘Chemo- our summary estimate by seeking evidence of publication bias,
therapy Induced Neurotoxicity’’ OR ‘‘Chemotherapy Induced selective outcome reporting bias (if a published protocol of the
Neurotoxicity Syndromes’’ OR ‘‘CIPN’’ OR ‘‘Oxaliplatin Induced included study was available), reporting of a sample size calcula-
Peripheral Neuropathy’’ OR ‘‘Bortezomib Induced Peripheral Neu- tion, and whether the study reported participants lost to follow-up.
ropathy’’ OR ‘‘Paclitaxel Induced Peripheral Neuropathy’’ OR ‘‘Tax-
ane Induced Peripheral Neuropathy’’ OR ‘‘Cisplatin Induced 2.4. Data synthesis and analysis
Peripheral Neuropathy’’ OR ‘‘Vincristine Induced Peripheral Neu-
ropathy’’ OR ‘‘Thalidomide Induced Peripheral Neuropathy’’ OR Our primary outcome was the prevalence of CIPN. We used ran-
‘‘Platinum Induced Peripheral Neuropathy’’ OR ‘‘Carboplatin dom effects meta-regression to quantify heterogeneity and its
Induced Peripheral Neuropathy’’ OR ‘‘Docetaxel Induced Periphe- potential sources. We hypothesized that chemotherapy type and
ral Neuropathy’’ OR ‘‘Proteasome Inhibitor Induced Peripheral the time of CIPN assessment would explain a large proportion of
Neuropathy’’ OR Neurotoxic Chemotherapy Induced Peripheral the observed heterogeneity. Therefore, we included chemotherapy
Neuropathy’’ OR ‘‘Cancer Neuropathic Pain’’ OR ‘‘Chemotherapy type, last time point of CIPN assessment, and measures of study
Induced Neuropathic Pain’’] [Search 1] AND [‘‘Prevalence’’ OR quality as independent variables in our regression model. We also
‘‘Epidemiology’’ OR ‘‘Occurrence’’ OR ‘‘Burden’’] [Search 2] AND planned for assessment of risk factors for CIPN across studies. We
[‘‘Predictors’’ OR ‘‘Risk Factors’’] [Search 3]. The search strategy assessed publication bias using funnel plots, Egger’s test, and trim
was adapted for each database (see supplementary text A). We and fill [22]. We appraised studies using STROBE criteria for obser-
also hand searched reference lists of relevant studies and system- vational studies and CONSORT criteria for trials. Where a criterion
atic reviews of CIPN prevention trials, and searched the databases was partially met, we considered, for the purposes of this analysis,
of National Institute for Health and Care Excellence (NICE) and that it was completely met, for ease of calculation. In open label
the Scottish Intercollegiate Guidelines Network (SIGN). Our studies (Table 1), we modified the CONSORT criteria by not consid-
review followed an a priori protocol according to the Preferred ering the point for blinding, to account for the design of these stud-
Reporting Items for Systematic Reviews and Meta-Analyses ies. STATA 13.1 was used for statistical analyses.
(PRISMA) guidelines [43]. The review protocol was registered on
the PROSPERO website before data extraction (registration no. 3. Results
CRD42013005524) [11].
3.1. Studies included
2.2. Inclusion and exclusion criteria and study selection
We identified 4128 potentially relevant studies, and examined
We included prospective observational studies of adult cancer the full text of 138. A total of 31 studies (involving 4179 patients)
patients receiving chemotherapy of any type. Our definition of [4–9,13,14,18,21,24–27,32–36,38,39,45–48,52,53,60,63–65] met
observational studies included cohort studies in which patients our inclusion criteria. A total of 30 studies reported the incidence
were prospectively identified and followed up using relevant pre- of CIPN (new CIPN cases divided by the population at risk). One
defined outcomes of interests. We also included control group data study reported CIPN prevalence (all CIPN cases divided by popula-
from randomized controlled trials (RCTs) of CIPN prevention in tion at risk) [26]. Because CIPN might have occurred, and resolved,
which details of the patients who developed CIPN were reported. between study assessments, we calculated the prevalence of CIPN
Studies were excluded if they described animal models of CIPN, at the time of each assessment [59].
were investigating CIPN treatment or prevention, included pediat-
ric populations, or investigated other causes of neuropathy in 3.2. Study characteristics
cancer patients (eg, pre-existing neuropathy such as diabetic
neuropathy or other cancer related causes of neuropathy such as Of the 31 studies included, 15 were prospective cohort studies,
post-mastectomy). 10 were RCTs, 5 were nonrandomized controlled trials, and 1 was a
Two investigators (M.S. and S.R.) independently read and cross-sectional cohort study. All nonrandomized controlled trials
selected from all the retrieved references and abstracts. Discrepan- were open labeled and not blinded. Eight of 10 RCTs (80%) reported
cies between the reviewers’ selections were resolved by discussion. investigator blinding of some type. Blinded assessment of outcome
Full texts of potentially eligible studies were retrieved (Fig. 1). was reported in 3 of 14 prospective cohort studies. One prospective
 Ò
M. Seretny et al. / PAIN 155 (2014) 2461–2470 2463

CIPN Incidence and Prevalence Systematic

Review and Meta-Analysis
PRISMA 2009 Flow Diagram

Records identified through database Additional records identified

searching through other sources
(n = 4109) (n = 19)

Records after duplicates removed

(n = 3657)

Records screened Records excluded

(n = 3657) (n = 3489)

Full-text articles assessed Full-text articles excluded,

for eligibility with reasons
(n = 168) (n = 137):

Inappropriate Design
(n = 83)
Studies included in
qualitative synthesis CIPN Treatment Study
(n = 31) (n = 1)

Incomplete Data Summary

(n = 48)

Studies included in Mixed Cancer Neuropathy

quantitative synthesis (n = 5)
(meta-analysis)
(n = 31)

Fig. 1. Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2009 flow diagram.

cohort study also sought to validate genetic risk factor results in a cumulative dose (CD) of chemotherapy (actual dose received) in
control group. Nine of 10 RCTs (90%) described a sample size calcu- our meta-regression because standard and maximally tolerated
lation. Of all included studies, 22 (71%) reported study participant doses would differ substantially from drug to drug (study-specific
dropout, giving reasons. In all, 14 of 31 study authors (45%) dis- CD shown in Table 1). As expected, there was co-linearity between
closed funders and/or whether they had a conflict of interest. the cancer type and the chemotherapy used; because we reasoned
Adherence of studies to reporting guidelines is summarized in that it is more likely that CIPN prevalence would be related to drug
Table 1. Of 31 studies, 26 (83.9%) used an assessment tool validated than to cancer type, we considered only chemotherapy type in our
for CIPN. All studies reporting CIPN risk factors described methods regression model (Table 3). The type of chemotherapy used
used to identify these predictors. accounted for 32% of the observed heterogeneity in our sample
(adjusted R2 = 0.315, P < .04).
3.3. CIPN incidence and prevalence Methods used to assess the presence or grade of CIPN were too
diverse to include in the meta-regression. Of the 31 included stud-
Of 4179 patients, 1960 developed CIPN (aggregate prevalence ies, 8 defined CIPN according to the National Cancer Institute Com-
48%). CIPN prevalence was 68.1% (95% CI = 57.7–78.4) within the mon Toxicity Criteria (NCI-CTC), 1 study used the European
first month of the end of chemotherapy, 60.0% (36.4–81.6) at Organisation for Research and Treatment of Cancer (EORTC) Qual-
3 months, and 30.0% (6.4–53.5) at 6 months or later (Table 2). ity of Life Questionnaire 30 (QLQ – 30) combined with neurological
There was considerable heterogeneity in the estimates from differ- examination, 1 used in-depth neurophysiological examination
ent studies (I2 = 98.2, P < .001). The time of assessment accounted (NPS), 1 used a standard neurological examination, and 1 used
for 36% of the observed heterogeneity (adjusted R2 = 0.365, the Total Neuropathy Score (TNSc). The remaining 18 studies used
P < .001). An overview of the individual incidence reported in a combination of 2 or more of the above, and 1 study used skin
included studies is shown in Table 1. We did not include the biopsy (Table 3). To investigate any impact of neurophysiological
 Ò
2464 M. Seretny et al. / PAIN 155 (2014) 2461–2470

Table 1
Overview of included studies.

First author (year) Study type and quality Incidence (95% CI) Main cancer class Dose (mg/m2)
(CONSORT/STROBE score) (chemotherapy) (mean or cumulative)
Antonacopoulou (2009)* Prospective cohort 58.8% (42.2–75.3) Colorectal (oxaliplatin) —
Argyriou (2006) Prospective cohort (18/22) 61.5% (35.1–87.9) Breast (paclitaxel) 1980
42.8% (16.9–68.7) Lung (cisplatin) 720
Argyriou (2007) [8] Prospective cohort (19/22) 64% (45.2–82.8) Colorectal (oxaliplatin) 1740
Argyriou (2007) Prospective cohort (19/22) 69.2% (44.1–94.3) Multiple solid (cisplatin and paclitaxel) 126
7
Argyriou (2012) Prospective cohort (19/22) 83.3% (77.3–89.3) Colorectal (oxaliplatin) 1646
Argyriou (2013) Prospective cohort (20/22) 84.5% (79.4–89.5) Colorectal (oxaliplatin) 1651
Attal (2009) Prospective cohort (19/22) 66.6% (44.8–88.4) Colorectal (oxaliplatin) 1278
Baldwin (2012) Prospective cohort (20/22) 67.2% (64.1–70.3) Breast (paclitaxel) —
Cascinu (1995) RCT (18/25) 64% (45.2–82.8) Gastrointestinal (cisplatin) —
Cascinu (2002) RCT (16/25) 78.9% (60.6–97.3) Colorectal (oxaliplatin) 783
 
Chaudhary (2008) Prospective cohort (13/22) 96.2% (89.2–103) Multiple myeloma (bortezomib and thalidomide) 36
Dimopoulos (2011) RCT (21/25) 46.7% (41.4–52.1) Multiple myeloma (bortezomib) 38
4
à
Gandara (1995) RCT (18/25) 12.1% (5.6–18
5) Ovarian and lung (cisplatin) 379
Ghoreishi (2012) RCT (19/25) 59.2% (40.7–77.8) Breast (paclitaxel) —
 
Glendenning (2010) Cross sectional cohort (21/22) 20.1% (15.5–24.7) Testicular (cisplatin and vincristine) 400
Gobran (2013) RCT (13/25) 70% (53.6–86.4) Colorectal (oxaliplatin) 763
Ishibashi (2010) RCT (20/25) 93.7% (81.9–105) Colorectal (oxaliplatin) 72
8
Johnson (2011) RCT (23/25) 32.1% (29.1–34.9) Multiple myeloma (thalidomide) —
19.6% (16.3–22.9) (Vincristine) —
 
Kawakami (2012) Prospective cohort (14/22) 76% (64.1–87.8) Lung (cisplatin and paclitaxel) —
Kemp (1996) RCT (19/25) 67
5% (59.2–75.8) Gynecological (cisplatin) —
Krishnan (2005) Prospective cohort (16/22) 50% (25.5–74.5) Colorectal (oxaliplatin) 1200
Lin (2006) Randomised trial (15/24) 90% (71.4–108) Colorectal (oxaliplatin) 1200
Milla (2009) Randomised trial (11/24) 92.8% (79.3–106) Colorectal (oxaliplatin) 772
Pace (2003) Randomised trial (11/24) 85.7% (67.4–104) Multiple solid (cisplatin) 420
Pace (2007) Prospective cohort (14/22) 92.8% (79.4–106) Breast (paclitaxel) 1744
Pace (2010) RCT (19/25) 41.6% (21.9–61.4) Multiple solid (cisplatin) 450
Planting (1999) Randomised trial (13/24) 13.5% (2.5–24.5) Multiple solid (cisplatin) 401
Plasmati (2002) Prospective cohort (15/22) 96% (88.3–103) Multiple myeloma (thalidomide) 18
Van der Hoop (1999) RCT (12/25) 41.6% (13.7–69.5) Gynecological (cisplatin) 416
Von Schlippe (2001) Prospective cohort (9/22) 17.2% (3.4–30.9) Testicular (cisplatin) —
Won (2012) Prospective cohort (16/22) 40.6% (30.8–50.4) Colorectal (oxaliplatin) 935

Abbreviation: RCT, randomized controlled trial (note that randomised trials, as opposed to RCTs, did not have blinding or placebo).
— Cumulative or average dose not reported. Reported cumulative dose refers to actual dose received.
*
Abstract only available; STROBE assessment not possible. Where upper 95% confidence intervals exceeded 100, only 100% were recorded, as this is clinically interpretable.
 
Study pooled incidence across chemotherapy types included.
à
Study pooled incidence across cancer types.

assessment on the reported prevalence of CIPN, we conducted a to validate candidate risk factors. Reported clinical risk factors
post hoc sensitivity analysis. In all, 17 studies (449 patients) used for CIPN included baseline neuropathy, a history of smoking,
NPS to assess for CIPN; 16 of these used NPS in combination with decreased creatinine clearance, and specific sensory changes dur-
another assessment method. In these 17 studies, CIPN prevalence ing chemotherapy treatment, including cold allodynia (pain in
was higher; 73.3% (58.6–87.3) within 1 month of chemotherapy response to a nonpainful cold stimulus) and cold hyperalgesia
cessation, 70.1% (41.8–98.4) at 3 months, and 39.9% (3.9–76.0) at (exaggerated pain in response to a painful cold stimulus, 20 °C).
6 months or more.
For publication bias, although Egger’s test did not suggest asym- 4. Discussion
metry in the funnel plot at a confidence level of P = .05 (95% CI of
intercept 0.64 to 7.8); trim and fill analysis did impute 14 4.1. CIPN prevalence
theoretical missing studies. These 2 approaches to assess for
publication bias are known to have different sensitivities [58]. This systematic review and meta-regression suggests a high
overall prevalence of CIPN, maximum within the first month after
3.4. CIPN risk factors treatment, and falling over time. Approximately one-third of
patients can expect to have chronic CIPN 6 months or more after
Eight of the included studies assessed risk factors for CIPN the end of chemotherapy; this has a significant negative impact
(Table 4) [8,9,21,26,33,34,48,65]. Four genome-wide association on long-term quality of life for which effective treatment is needed.
studies (GWAS), totaling 2671 patients, sought single nucleotide The lack of uniformity in CIPN assessment methods make
polymorphisms (SNPs) associated with CIPN [9,33,48,65]. All between-study comparisons difficult. Authors used 5 assessment
GWAS used validation datasets and conducted genotyping blinded methods (NCI-CTC, TNSc, EORTC QLQ-C30, neuro-physiological
to clinical status. These reported polymorphisms associated with a examination, which included nerve conduction studies and/or
range of proteins, including voltage-gated sodium channels, Schw- quantitative sensory testing, and neurological examination) alone
ann cell function–related proteins, receptors for cell surface colla- or in combination. Of these, only the EORTC QLQ-C30 and quanti-
gen, receptors involved in neuronal apoptosis, neuronal crest cell tative sensory testing component of neurophysiological examina-
development, and an enzyme involved in pyruvate metabolism. tion explicitly assess pain as a symptom of CIPN. It is known that
Four studies (701 patients) used statistical modeling to report although CIPN most frequently presents with pain, motor and
clinical risk factors for CIPN [8,21,26,34]. Two of these studies other sensory symptoms may also be present [40]. Use of combina-
included 50 patients or fewer. No study used a separate data set tions of CIPN and pain assessment tools has been suggested as a
 Ò
M. Seretny et al. / PAIN 155 (2014) 2461–2470 2465

Table 2
Comparison of prevalence related to time of CIPN assessment.

Time of assessment Prevalence (95% CI) Studies included Total no. of patients in group
(after cessation of chemotherapy)
61 mo 68.1% (57.7–78.4) Antonacopolou 2009 2085
Argyriou 2007
Argyriou 2012
Argyriou 2013
Baldwin 2012
Cascinu 1995
Cascinu 2002
Chaudhry 2008
Dimopoulos 2011*
Gandara 1995
Ghoreishi 2012
Gobran 2013*
Ishibashi 2010
Kawakami 2012
Krishnan 2005*
Lin 2006
Milla 2009*
Pace 2003
Pace 2007*
Pace 2010
Van Der Hoop 1999
Won 2012
 
3 mo 60.0% (36.4–81.6) Argyriou 2006 234
Argyriou 2007
Kemp 1996
Planting 1999
Plasmati 2007
  à
P6 mo 30.0% (6.4–53.5) Johnson 2011 1860
Attal 2009
Glendenning 2010
Von Schlippe 2001

Abbreviations: CI, confidence interval; CIPN, chemotherapy-induced peripheral neuropathy.

*
Studies included longer-term CIPN follow up but did not provide enough details at these later time points to allow use of data in the meta-
regression.
 
Wide confidence interval likely due to small number of studies assessing CIPN beyond this time point.
à
Study considered CIPN only after induction therapy and not during maintenance.

strategy to improve detection and quantification of pain in CIPN 4.2. Risk factors for CIPN
[67]. There have been recent attempts to standardize CIPN assess-
ment and reporting, and we encourage investigators to consider Four studies used multivariate statistical modeling to identify
these when developing study protocols [15,16]. clinical risk factors for CIPN [8,21,26,34]. Despite using valid statis-
Three of the 5 largest studies in our sample did not include the tical approaches, these studies did not verify identified risk factors
mildest grades of CIPN [9,24,45]. The prevalence of CIPN is there- in new population datasets. Consequently, their results are proba-
fore likely to be higher than reported here. Early detection of mild bly affected by the statistical biases underpinning these types of
CIPN might become important if effective prevention or manage- predictive calculations [3,42]. To our knowledge, these are the only
ment strategies become available. A lower incidence in these larger studies that describe baseline neuropathy, smoking, and decreased
studies is an alternative explanation for the funnel plot asymmetry creatinine clearance as risk factors for CIPN. In contrast, description
detected by trim and fill analysis [58]. of sensory changes during chemotherapy treatment, including
Current clinical guidelines support use of NPS methods in the increased pain and nerve hyperexcitability, have previously been
diagnosis of suspected CIPN [19,56]. Studies using this approach documented as predictors of CIPN [20,42]. The postulated mecha-
reported a higher prevalence of CIPN, but whether this is a clini- nisms underpinning these sensory phenomena include axonal
cally significant problem is not clear. hyperexcitability and nociceptor sensitization. These processes
We found significant heterogeneity between studies. In meta- may be important in CIPN development, and, to some degree, they
analyses aimed at providing a best estimate of, for instance, drug fit with the mechanisms described in other neuropathic conditions
efficacy, significant heterogeneity usually limits the usefulness of related to systemic diseases, including human immunodeficiency
pooled data. In contrast, because the etiology and epidemiology virus (HIV) and multiple sclerosis [42,64]. There is ongoing debate
of CIPN are so poorly understood, we believe that investigating about the relative importance of etiology in determining the
the sources of heterogeneity is important. Specifically, it might underlying mechanisms of neuropathic pain [19,56,62].
provide insight into the impact of length of assessment and chemo- Four studies reported genetic risk factors for CIPN. The functions
therapy type on the incidence and prevalence of CIPN. Further- of the identified genes fit with the postulated pathophysiological
more, as expected, a substantial proportion of the heterogeneity mechanisms underpinning CIPN [50]. The recent comprehensive
that we observed was accounted for by chemotherapy type, which review by Cavaletti et al. discusses these mechanisms in detail.
was related to the cancer type. Although the primary interest of All 4 included studies were, to some degree, affected by the univer-
many clinicians will be the prevalence of CIPN for specific chemo- sal limitations influencing pharmacogenetic studies: inadequate
therapeutics, CIPN treatment decisions are routinely based on data sample size, CIPN assessment tools, and use and size of a replication
from treatment trials that have recruited patients irrespective of cohort. Despite these possible limitations, the potential clinical use-
the chemotherapy that they were prescribed [57]. fulness of pharmacogenetic studies in CIPN has recently been
 Ò
2466 M. Seretny et al. / PAIN 155 (2014) 2461–2470

Table 3
Studies stratified by drug type.

Study type Main cancer CIPN severity report CIPN assessment time points CIPN assessment
(CONSORT/STROBE) class (count by grade if given) method(s)
Oxaliplatin: 72.3% (95% CI = 59.7–86.8)
Antonacopoulou Prospective cohort Colorectal NR Unclear TNSc
(2009)*
Argyriou (2007) [8] Prospective cohort Colorectal Grade I (6/16) Baseline TNSc
Grade II (8/16) Cycles 4, 8, 12 NPS
Grade III (2/16) NCI-CTC
Argyriou (2012) Prospective cohort Colorectal Grade I (38/125) Baseline TNSc
Grade II (46/125) Cycles 3, 6 (FOLFOX) NPS
Grade III (41/125) Cycles 4, 8 (XELOX) NCI-CTC
 
Argyriou (2013) Prospective cohort Colorectal Grade I (62/169) Baseline TNSc
Grade II (46/169) Cycle 6, 12 (FOLFOX) NCI-CTC
Grade III (61/169) Cycles 4, 8 (XELOX)
Attal (2009) Prospective cohort Colorectal Sensory symptom counts Baseline NCI-CTC
described as means/ Cycle 3, 6, 9 NPS
individual 12 ± 2 mo after chemo end (EORTC) QLQ-C30
Cascinu (2002) RCT Colorectal Grade I (4/15) Baseline NCI-CTC
Grade II (6/15) Cycles 4, 8, 12 NPS
Grade III (4/15) Within 2 wk of chemo end
Grade IV (1/15)
Gobran (2013) RCT Colorectal Grade I (7/21) Unclear if at baseline NCI-CTC
Grade II (0/21) At each chemo cycle until end of chemo
(variable no. of cycles)
Grade III (14/21) Longer follow-up for those with CIPN (but
denominator unclear)
Grade IV (0/21)
Ishibashi (2010) RCT Colorectal Grade I (15/15) Baseline NCI-CTC
Grade II (1/15) At each chemo cycle until end of chemo
Grade III (0/15)
Grade IV (0/15)
Krishnan (2005) Prospective cohort Colorectal NR No baseline NCI-CTC
Within 1 mo of chemo end only reported NPS
assessment
TNSc
Lin (2006) Controlled trial Colorectal Grade I (1/9) Baseline NCI-CTC
Grade II (5/9) Cycles 4, 8, 12 NPS
Grade III (3/9) Within 2 wk of end of chemo
Grade IV (0/9)
Milla (2009) Controlled trial Colorectal Grade I (0/13) Baseline NCI-CTC
Grade II (9/13) Cycles 5, 9, 12 NES
Grade III (4/13) (Some followed up longer but denominator
unclear)
Won (2012) Prospective cohort Colorectal NR Unclear if at baseline NCI-CTC
At each chemo cycle until end of chemo NES
(variable no. of cycles)
Cisplatin: 42.2% (95% CI = 21.3–63.1)
à
Argyriou (2006) Prospective cohort Lung Reported by age group Baseline PNS
only Cycles 3, 6 NPS
3 mo after chemo end
Cascinu (1995) RCT Gastrointestinal Grade I (3/16) Baseline NCI-CTC
Grade II (10/16) After 9 and 15 wk of therapy NPS
Grade III (2/16) Within 1 wk after end of chemo
Grade IV (1/16)
Gandara (1995) RCT Ovarian and Only grade P3 reported Unclear if at baseline NCI-CTC
lung At each cycle until chemo end (variable no. of
cycles)
Study stopped early after interim analysis due
to high toxicity in intervention group
Kemp (1996) RCT Gynecological Grade I (31/81) Baseline NCI-CTC
Grade II (35/81) Cycles 4, 5, 6 NES
Grade III (15/81) Monthly after chemo for 3 mo
Pace (2003) Controlled trial Multiple solid Grade I (6/12) Baseline TNSc
Grade II (4/12) After 6 cycles NES
Grade III & IV (2/12)
Pace (2010) RCT Multiple solid Only grade P3 reported Baseline TNSc
Every cycle for 3 cycles NPS
1 mo after chemo end
Planting (1999) Controlled trial Multiple solid Grade I (5/5) Baseline NCI-CTC
Cycle 3, 6 NES
3 mo after chemo end
(Longer follow-up but no denominator info)
Van der Hoop (1999) Controlled trial Gynecological Mean vibration threshold Baseline NES
Cycles 2, 4, 6
End of chemo
 Ò
M. Seretny et al. / PAIN 155 (2014) 2461–2470 2467

Table 3 (continued)

Study type Main cancer CIPN severity report CIPN assessment time points CIPN assessment
(CONSORT/STROBE) class (count by grade if given) method(s)
Von Schlippe (2001) Prospective cohort Testicular Grade I (4/5) Unclear if at baseline NPS
Grade II (1/5) Every 6 wk for first 6 mo after chemotherapy
Thereafter every 2 mo for median of 4 y (range
2–8 y)
Cisplatin or carboplatin and paclitaxel: 73% (95% CI = 36.2–109.7)
Argyriou (2007) Prospective cohort Multiple solid Mild (2/9) Baseline PNS
Moderate (6/9) Cycle 3, 6 NPS
Severe (1/9) 3 mo after chemo end
Kawakami (2012)§ Prospective cohort Lung % Severity with Baseline NCI-CTC
cumulative dose Daily during cycle 1
Cycle 2, 3, 4
Chemo end
Cisplatin and vincristine: 20.1% (95% CI = 26.2 to 66.5)
Glendenning (2010)§ Cross-sectional Testicular Only grade P3 reported Recruited patients at least 5 y post-treatment (EORTC) QLQ-C30
cohort Assessed once for this prevalence study NES
Paclitaxel: 70.8% (95% CI = 43.5–98.1)
à
Argyriou (2006) Prospective cohort Breast Reported by age group Baseline PNS
only Cycles 3, 6 NPS
3 mo after chemo end
Baldwin (2012) Prospective cohort Breast Only grade P2 reported Unclear if at baseline NCI-CTC
Cycles 4, 6
Within 1 mo of chemo end
Ghoreishi (2012) RCT Breast Mild (10/16) Baseline TNSc
Moderate (5/16) 1 mo after chemo end NPS
Severe (1/16)
Pace (2007) Prospective cohort Breast Mean neurotoxicity scores Baseline TNSc
reported After 12 wk of chemo NPS
After 24 wk of chemo
Vincristine: 19.6% (95% CI 26.6 to 65.9)
à
Johnson (2011) RCT Multiple Grade P I 31.8% Unclear if at baseline NCI-CTC
myeloma Grade P II 11% At each cycle
Grade P III 3.6% For 6 months after chemo end for induction (ie,
36 wk from start of induction therapy)
Thalidomide: 63.5% (95% CI = 29.3–97.8)
à
Johnson (2011) RCT Multiple Grade details not reported Unclear if at baseline NCI-CTC
myeloma At each cycle
For 6 mo after end of chemo for induction (ie,
36 weeks from start of induction therapy)
Plasmati (2002) Prospective cohort Multiple Grade I (12/24) Baseline NCI-CTC
myeloma Grade II (6/24) After 4 mo of chemo NPS
Subclincial (6/24) 3 mo after stem cell transplantation
Bortezomib: 46.7% (95% CI = 0.3–93.1)
Dimopoulos (2011) RCT Multiple Grade I NR Unclear if at baseline NCI-CTC
myeloma Grade II (64/159) Every 3 wk until
Grade III (45/159) 1 mo after last chemo dose
Grade IV (1/159) Longer follow-up but no denominator data
Bortezomib and thalidomide: 96.2% (95% CI = 49.7–143)
Chaudhary (2008) Prospective cohort Multiple Grade P2 reported Baseline TNSc
myeloma Cycles 2, 4, 6, 8 NPS
End of chemo Skin biopsy
Note skin biopsy at baseline and end of chemo
only

Abbreviations: Chemo, chemotherapy; CIPN, chemotherapy-induced peripheral neuropathy; EORTC, European Organization for Research and Treatment of Cancer; CI,
confidence interval; NCT-CTC, National Cancer Institute Common Toxicity Criteria; NES, neurological examination; NPS, neurophysiological examination (quantitative
sensory testing and/or nerve conduction studies); NR, not reported; PNS, Modified peripheral neuropathy score; RCT, randomized controlled trial; TNSc, total neuropathy
score.
*
Abstract only available.
 
Authors report both acute and chronic CIPN grade counts, only acute given here.
à
Raw data obtained from author or reported in paper, allowing counts reported in single study to be split by chemotherapy type.
§
Studies pooled CIPN counts across chemotherapy types included.

described [10]. As suggested by Postma et al. adherence of future non–English language studies. Multivariate meta-regression would
studies to standardized study design and methods will likely aid have allowed us to investigate interactions between various
the advance of personalized oncology, possibly having an impact factors, but there are too few studies for this approach to be
on CIPN prevalence in the future. reliable. Because we expected there to be a broad range of CIPN
assessment methods used, we did not plan to explore their
4.3. Limitations of this review impact. Our analysis of the impact of NPS as a component of
the assessment of CIPN is post hoc and therefore should be
It is possible that we have omitted relevant studies despite interpreted with caution. We did not specifically seek out
our detailed search strategy, and we specifically excluded assessments for pain in CIPN in included studies and therefore
 Ò
2468 M. Seretny et al. / PAIN 155 (2014) 2461–2470

Table 4
CIPN risk factors.

Study Category of risk factor reported Data source of study Sample size of study (N) Risk factor details
Argyriou (2013) Genetic Prospective cohort 200 SNC4A-rs2302237 OR = 2.65 (1.15–6)
SCN10A-rs1263292 OR = 0.39 (0.17–0.88)
Attal (2009) Clinical Prospective cohort 18 Cold allodynia OR = 39 (1.8–817)
Cold hyperalgesia OR = 3.9 (1.0–1.20)
Baldwin (2012) Genetic Prospective cohort 855 FGD4-rs10771973 HR = 1.57 (1.30–1.91)
Dimopoulos (2011) Clinical RCT 340 Baseline neuropathy HR = 1.79 (p < 0.01)
Glendenning (2010) Clinical and treatment-related Cross-sectional cohort 293 Cisplatin dose increase OR = 1.91 (1.61–2.26)
Carboplatin dose increase OR = 1.26 (1.04–1.52)
Age at follow-up OR = 1.06 (1.04–1.08)
⁄
Johnson (2011) Genetic RCT 970 + 550 ABCA1-rs363717 OR = 0.71 (0.52–0.98)
ICAM1-rs1799969 OR = 0.67 (0.44–1.03)
PPARD-rs2076169 OR = 0.60 (0.38–095)
SERPINB2-rs6103 OR = 0.70 (0.52–0.95)
SLC12A6-rs7164902 OR = 0.60 (0.44–0.80)
Kawakami (2012) Clinical Prospective cohort 50 Smoking history pack-years HR = 1.03 (1.0–1.05)
Decreased creatinine clearance HR = 0.96 (0.92–0.99)
 
Won (2012) Genetic Prospective cohort 96 TAC1-rs10486003
FOXC1-rs2338
ITGA1-rs830884
ACYP2-rs843748
DLEU7-rs797519

Abbreviations: CIPN, chemotherapy-induced peripheral neuropathy; HR, hazard ratio (95% confidence interval or significance level); OR, odds ratio (95% confidence interval);
RCT, randomized controlled trial; SNP, single nucleotide polymorphism.
Note that Jonson et al. reported ORs for both populations included in their analysis. Only 1 set of ORs is reported here. All effect sizes reported here are directly from the cited
studies.
*
SNP association with CIPN grade P2 only.
 
Won et al. reported the overall predictive accuracy of the multiple logistic regression model yielding the 5 positive single nucleotide polymorphisms (SNPs), 72.8% (65.8–
79.9), as opposed to ORs for individual SNPs.

are unable to quantify prevalence of painful CIPN explicitly in Acknowledgements

out analysis.
The authors thank Marshall Dozier, Librarian, University of
4.4. Strengths of this review Edinburgh, for her help explaining techniques regarding system-
atic review search strategies. M.S. is funded by the Wellcome Trust,
Our meta-analysis quantifies CIPN prevalence across most che- through the Scottish Translational Medicine and Therapeutics Ini-
motherapy and cancer types. This allows our prevalence measures tiative (STMTI). G.C. is funded by the National Centre for the
to be used by clinicians when deciding between chemotherapy Replacement, Refinement and Reduction of Animals in Research
types and regimens. It is also useful for planning future CIPN treat- (NC3Rs). E.S. is funded by the Seventh Framework Programme –
ment studies. In addition, these findings may be useful for both European Union and the NC3Rs. M.ML. is funded by the Seventh
resource allocation and research planning. Our pooled prevalence Framework Programme and the NC3Rs. S.R. is supported by the
also allows direct estimation of economic costs of CIPN resulting Melville Trust. R.G. is a NHS consultant and honorary lecturer at
from the chemotherapeutics and cancer types included in our the University of Edinburgh. L.C. is a reader at the University of
review [51]. Edinburgh and honorary NHS consultant. M.F. holds the St
In this first meta-analysis investigating epidemiological mea- Columba’s Hospice Chair of Palliative Medicine at the University
sures of CIPN, we highlight the effect of the time of assessment, of Edinburgh. Funders were not involved in any part of the design,
after chemotherapy cessation, on CIPN prevalence. This has impli- execution, or interpretation of this study. The corresponding
cations for surveillance of CIPN at follow up, clinical care planning, author had full access to all the data in the study and had final
and patient expectations. Specifically, our results may contribute responsibility for the decision to submit for publication.
to explaining the risks of developing CIPN, and its likely natural
history, to patients at consent for chemotherapy. In broad terms, Appendix A. Supplementary data
around two-thirds of patients will suffer from CIPN in the first
month after chemotherapy, but in only one-half of these will CIPN Supplementary data associated with this article can be found, in
have resolved by six months. Finally, we have confirmed the urgent the online version, at http://dx.doi.org/10.1016/j.pain.2014.09.020.
need for a standardized approach to the diagnosis of CIPN, reaf-
firming ongoing efforts such as those of the chemotherapy-induced References
peripheral neuropathy outcome measures standardization study
(CI-PERINOMS) group [67]. [1] Albers JW, Chaudhry V, Cavaletti G, Donehower CR. Interventions for
preventing neuropathy caused by cisplatin and related compounds. Cochrane
Database Syst Rev 2011;16:CD005228.
Conflict of interest statement [2] Altman DG, Schulz KF, Moher D, Egger M, Davidoff F, Elbourne D, Gotzsche PC,
Lang T, for the CONSORT Group. The revised CONSORT statement for reporting
randomized trials: explanation and elaboration. Ann Intern Med
Marta Seretny, Gillian Currie, Emily Sena, Malcolm MacLeod, 2001;134:663–94.
Robin Grant, and Marie Fallon declare no conflicts of interest. Les- [3] Altman DG, Vergouwe Y, Royston P, Moons KGM. Prognosis and prognostic
ley Colvin serves as an editor for the British Journal of Anaesthesia. research: validating a prognostic model. BMJ 2009;338:b605.
 Ò
M. Seretny et al. / PAIN 155 (2014) 2461–2470 2469

[4] Argyriou AA, Polychronopoulos P, Iconomou G, Koutras A, Makatsoris T, [27] Gobran NS. Role of calcium and magnesium infusion in prevention of
Gerolymos MK, Gourzis P, Assimakopoulos K, Kalofonos HP, Chroni E. oxaliplatin neurotoxicity. A phase III trial. Chinese-German J Clin Oncol
Incidence and characteristics of peripheral neuropathy during 2013;12:232–6.
oxaliplatin-based chemotherapy for metastatic colon cancer. Acta Oncol [28] Grisold W, Cavaletti G, Windebank AJ. Peripheral neuropathies from
2007;46:1131–7. chemotherapeutics and targeted agents: diagnosis, treatment, and
[5] Argyriou AA, Polychronopoulos P, Koutras A, Iconomou G, Gourzis P, prevention. Neuro-Oncology 2012;14:45–54.
Assimakopoulos K, Kalofonos HP, Chroni E. Is advanced age associated with [29] Guyatt GH, Sinclair J, Cook DJ, Glasziou P. Users’ guides to the medical
increased incidence and severity of chemotherapy-induced peripheral literature: XVI. How to use a treatment recommendation. Evidence-Based
neuropathy? Support Care Cancer 2006;14:223–9. Medicine Working Group and the Cochrane Applicability Methods Working
[6] Argyriou AA, Polychronopoulos P, Koutras A, Xiros N, Petsas T, Argyriou K, Group. JAMA 1999;281:1836–43.
Kalofonos HP, Chroni E. Clinical and electrophysiological features of peripheral [30] Hammack JE, Michalak JC, Loprinzi CL, Sloan JA, Novotny PJ, Soori GS, Tirona
neuropathy induced by administration of cisplatin plus paclitaxel-based MT, Rowland KM, Stella PJ, Johnson JA. Phase III evaluation of nortriptyline for
chemotherapy. Eur J Cancer Care 2007;16:231–7. alleviation of symptoms of cis-platinum-induced peripheral neuropathy.
[7] Argyriou AA, Velasco R, Briani C, Cavaletti G, Bruna J, Alberti P, Cacciavillani M, PAINÒ 2002;98:195–203.
Lonardi S, Santos C, Cortinovis D, Cazzaniga M, Kalofonos HP. Peripheral [31] Hershman DL, Lacchetti C, Dworkin RH, Smith EML, Bleeker J, Cavaletti G,
neurotoxicity of oxaliplatin in combination with 5-fluorouracil (FOLFOX) or Chauhan C, Gavin P, Lavino A, Lustberg MB, Paice J, Schneider B, Smith ML,
capecitabine (XELOX): a prospective evaluation of 150 colorectal cancer Smith T, Terstriep S, Wagner-Johnston N, Bak K, Loprinzi CL. Prevention and
patients. Ann Oncol 2012;23:3116–22. management of chemotherapy-induced peripheral neuropathy in survivors of
[8] Attal N, Bouhassira D, Gautron M, Vaillant JN, Mitry E, Lepere C, Rougier P, adult cancers: American Society of Clinical Oncology Clinical Practice
Guirimand F. Thermal hyperalgesia as a marker of oxaliplatin neurotoxicity: a Guideline. J Clin Oncol 2014;32:1941–67.
prospective quantified sensory assessment study. PAINÒ 2009;144:245–52. [32] Ishibashi K, Okada N, Miyazaki T, Sano M, Ishida H. Effect of calcium and
[9] Baldwin RM, Owzar K, Zembutsu H, Chhibber A, Kubo M, Jiang C, Watson D, magnesium on neurotoxicity and blood platinum concentrations in patients
Eclov RJ, Mefford J, McLeod HL, Friedman PN, Hudis CA, Winer EP, receiving mFOLFOX6 therapy: a prospective randomized study. Int J Clin Oncol
Jorgenson EM, Witte JS, Shulman LN, Nakamura Y, Ratain MJ, Kroetz DL. 2010;15:82–7.
A genome-wide association study identifies novel loci for paclitaxel-induced [33] Johnson DC, Corthals SL, Walker BA, Ross FM, Gregory WM, Dickens NJ,
sensory peripheral neuropathy in CALGB 40101. Clin Cancer Res Lokhorst HM, Goldschmidt H, Davies FE, Durie BGM, Van Ness B, Child JA,
2012;18:5099–109. Sonneveld P, Morgan GJ. Genetic factors underlying the risk of thalidomide-
[10] Boland EG, Selvarajah D, Hunter M, Ezaydi Y, Tesfaye S, Ahmedzai SH, related neuropathy in patients with multiple myeloma. J Clin Oncol
Snowden JA, Wilkinson ID. Central pain processing in chronic chemotherapy- 2011;29:797–804.
induced peripheral neuropathy: a functional magnetic resonance imaging [34] Kawakami K, Tunoda T, Takiguchi T, Shibata K, Ohtani T, Kizu J, Nishio M, Horai
study. PLoS One 2014;9:5. T, Hama T, Taguchi K. Factors exacerbating peripheral neuropathy induced by
[11] Booth A. PROSPERO: International register of systematic reviews, Available at: paclitaxel plus carboplatin in non-small cell lung cancer. Oncol Res
http://www.crd.york.ac.uk/PROSPERO; 2013 [accessed August 29, 2013]. 2012;20:179–85.
[12] Cancer Research UK. Cancer incidence and survival rates, Available at: http:// [35] Kemp G, Rose P, Lurain J, Berman M, Manetta A, Roullet B, Homesley H,
www.cancerresearchuk.org/cancer-info/cancerstats/keyfacts/Allcancerscom- Belpomme D, Glick J. Amifostine pretreatment for protection against
bined/; 2013 [accessed November 26, 2013]. cyclophosphamide-induced and cisplatin-induced toxicities: results of a
[13] Cascinu S, Catalano V, Cordella L, Labianca R, Giordani P, Baldelli AM, Beretta randomized controlled trial in patients with advanced ovarian cancer. J Clin
GD, Ubiali E, Catalano G. Neuroprotective effect of reduced glutathione on Oncol 1996;14:2101–12.
oxaliplatin-based chemotherapy in advanced colorectal cancer: a randomized, [36] Krishnan AV, Goldstein D, Friedlander M, Kiernan MC. Oxaliplatin-induced
double-blind, placebo-controlled trial. J Clin Oncol 2002;20:3478–83. neurotoxicity and the development of neuropathy. Muscle Nerve
[14] Cascinu S, Cordella L, Delferro E, Fronzoni M, Catalano G. Neuroprotective 2005;32:51–60.
effect of reduced glutathione on cisplatin-based chemotherapy in advanced [37] Lema MJ, Foley KM, Hausheer FH. Types and epidemiology of cancer-related
gastric-cancer—a randomized double-blind placebo-controlled trial. J Clin neuropathic pain: the intersection of cancer pain and neuropathic pain.
Oncol 1995;13:26–32. Oncologist 2010;15:3–8.
[15] Cavaletti G, Alberti P, Marmiroli P. Chemotherapy-induced peripheral [38] Lin PC, Lee MY, Wang WS, Yen CC, Chao TC, Hsiao LT, Yang MH, Chen PM, Lin
neurotoxicity in the era of pharmacogenomics. Lancet Oncol KP, Chiou TJ. N-acetylcysteine has neuroprotective effects against oxaliplatin-
2011;12:1151–61. based adjuvant chemotherapy in colon cancer patients: preliminary data.
[16] Cavaletti G, Cornblath DR, Merkies IS, Postma TJ, Rossi E, Frigeni B, Alberti P, Support Care Cancer 2006;14:484–7.
Bruna J, Velasco R, Argyriou AA, Kalofonos HP, Psimaras D, Ricard D, Pace A, [39] Milla P, Airoldi M, Weber G, Drescher A, Jaehde U, Cattel L. Administration of
Galie E, Briani C, Dalla Torre C, Faber CG, Lalisang RI, Boogerd W, Brandsma D, reduced glutathione in FOLFOX4 adjuvant treatment for colorectal cancer:
Koeppen S, Hense J, Storey D, Kerrigan S, Schenone A, Fabbri S, Valsecchi MG, effect on oxaliplatin pharmacokinetics, Pt-DNA adduct formation, and
CI-PeriNomS Group. The Chemotherapy-Induced Peripheral Neuropathy neurotoxicity. Anti-Cancer Drugs 2009;20:396–402.
Outcome Measures Standardization Study: from consensus to the first [40] Miltenburg NC, Boogerd W. Chemotherapy-induced neuropathy: a
validity and reliability findings. Ann Oncol 2013;24:454–62. comprehensive survey. Cancer Treat Rev 2014;40:872–82.
[17] Cavaletti G, Marmiroli P. Chemotherapy-induced peripheral neurotoxicity. Nat [41] Mitchell P, Goldstein D, Michael M, Beale P, Friedlander M, Zalcberg J, Clarke S,
Rev Neurol 2010;6:657–66. White S. Addition of gabapentin (G) to a modified FOLFOX regimen does not
[18] Chaudhry V, Cornblath DR, Polydefkis M, Ferguson A, Borrello I. Characteristics reduce neurotoxicity in patients (pts) with advanced colorectal cancer (CRC). J
of bortezomib- and thalidomide-induced peripheral neuropathy: research Clin Oncol 2005;23. 266S–266S.
report. J Peripher Nerv Syst 2008;13:275–82. [42] Moalem G, Tracey DJ. Immune and inflammatory mechanisms in neuropathic
[19] Colvin LA, Dougherty PM. Peripheral neuropathic pain: signs, symptoms, pain. Brain Res Rev 2006;51:240–64.
mechanisms, and causes: are they linked? Br J Anaesth 2014. Sep 24. pii: [43] Moher D, Liberati A, Tetzlaff J, Altman DG, PRISMA Group. Preferred reporting
aeu323. [Epub ahead of print]. items for systematic reviews and meta-analyses: the PRISMA statement. BMJ
[20] Costigan M, Scholz J, Woolf CJ. Neuropathic pain: a maladaptive response of 2009;339:b2535.
the nervous system to damage. Annu Rev Neurosci 2009;32:1–32. [44] National Cancer Institute. Cancer Survival Statistics, Available at: http://
[21] Dimopoulos MA, Mateos M-V, Richardson PG, Schlag R, Khuageva NK, seer.cancer.gov/statfacts/html/all.html; 2013 [accessed December 12, 2013].
Shpilberg O, Kropff M, Spicka I, Palumbo A, Wu KL, Esseltine D-L, Liu K, [45] Pace A, Giannarelli D, Galiè E, Savarese A, Carpano S, Della Giulia M, Pozzi A,
Deraedt W, Cakana A, van de Velde H, San Miguel JF. Risk factors for, and Silvani A, Gaviani P, Scaioli V, Jandolo B, Bove L, Cognetti F. Vitamin E
reversibility of, peripheral neuropathy associated with bortezomib– neuroprotection for cisplatin neuropathy: a randomized, placebo-controlled
melphalan–prednisone in newly diagnosed patients with multiple myeloma: trial. Neurology 2010;74:762–6.
subanalysis of the phase 3 VISTA study. Eur J Haematol 2011;86:23–31. [46] Pace A, Nistico C, Cuppone F, Bria E, Galie E, Graziano G, Natoli G, Sperduti I,
[22] Egger M, Smith GD, Schneider M, Minder C. Bias in meta-analysis detected by a Jandolo B, Calabretta F, Tomao S, Terzoli E. Peripheral neurotoxicity of weekly
simple, graphical test. Br Med J 1997;315:629–34. paclitaxel chemotherapy: a schedule or a dose issue? Clin Breast Cancer
[23] Fallon MT. Neuropathic pain in cancer. Br J Anaesth 2013;111:105–11. 2007;7:550–4.
[24] Gandara DR, Nahhas WA, Adelson MD, Lichtman SM, Podczaski ES, Yanovich S, [47] Pace A, Savarese A, Picardo M, Maresca V, Pacetti U, Del Monte G, Biroccio A,
Homesley HD, Braly P, Ritch PS, Weisberg SR, Williams L, Diasio RB, Perez EA, Leonetti C, Jandolo B, Cognetti F, Bove L. Neuroprotective effect of vitamin E
Karp D, Reich SD, McCarroll K, Hoff JV. Randomized placebo-controlled supplementation in patients treated with cisplatin chemotherapy. J Clin Oncol
multicenter evaluation of diethyldithiocarbamate for chemoprotection 2003;21:927–31.
against cisplatin-induced toxicities. J Clin Oncol 1995;13:490–6. [48] Pachman DR, Barton DL, Watson JC, Loprinzi CL. Chemotherapy-induced
[25] Ghoreishi Z, Esfahani A, Djazayeri A, Djalali M, Golestan B, Ayromlou H, peripheral neuropathy: prevention and treatment. Clin Pharmacol Ther
Hashemzade S, Asghari Jafarabadi M, Montazeri V, Keshavarz SA, Darabi M. 2011;90:377–87.
Omega-3 fatty acids are protective against paclitaxel-induced peripheral [49] Park SB, Goldstein D, Krishnan AV, Lin CS, Friedlander ML, Cassidy J,
neuropathy: a randomized double-blind placebo controlled trial. BMC Cancer Koltzenburg M, Kiernan MC. Chemotherapy-induced peripheral
2012;12:355. neurotoxicity: a critical analysis. CA-A Cancer J Clinicians 2013;63:419–37.
[26] Glendenning JL, Barbachano Y, Norman AR, Dearnaley DP, Horwich A, Huddart [50] Park SB, Krishnan AV, Lin CSY, Goldstein D, Friedlander M, Kiernan MC.
RA. Long-term neurologic and peripheral vascular toxicity after chemotherapy Mechanisms underlying chemotherapy-induced neurotoxicity and the
treatment of testicular cancer. Cancer 2010;116:2322–31. potential for neuroprotective strategies. Curr Med Chem 2008;15:3081–94.
 Ò
2470 M. Seretny et al. / PAIN 155 (2014) 2461–2470

[51] Pike CT, Birnbaum HG, Muehlenbein CE, Pohl GM, Natale RB. Healthcare costs Schwarzer G, Duval S, Altman DG, Moher D, Higgins JPT. Recommendations for
and workloss burden of patients with chemotherapy-associated peripheral examining and interpreting funnel plot asymmetry in meta-analyses of
neuropathy in breast, ovarian, head and neck, and nonsmall cell lung cancer. randomised controlled trials. Br Med J 2011;343:d4002.
Chemother Res Pract 2012;2012:913848. [59] Turner EL, Sweeting MJ, Lindfield RJ, DeAngelis D. Incidence estimation using a
[52] Planting AST, Catimel G, de Mulder PHM, de Graeff A, Hoppener F, Verweij J, single cross-sectional age-specific prevalence survey with differential
Oster W, Vermorken JB, for the EORTC Head and Neck Cooperative Group. mortality. Stat Med 2013;33:422–35.
Randomized study of a short course of weekly cisplatin with or without [60] Vanderhoop RG, Vecht CJ, Vanderburg MEL, Elderson A, Boogerd W, Heimans
amifostine in advanced head and neck cancer. Ann Oncol 1999;10:693–700. JJ, Vries EP, Vanhouwelingen JC, Jennekens FGI, Gispen WH, Neijt JP.
[53] Plasmati R, Pastorelli F, Cavo M, Petracci E, Zamagni E, Tosi P, Cangini D, Prevention of cisplatin neurotoxicity with an ACTH(4–9) analog in patients
Tacchetti P, Salvi F, Bartolomei I, Michelucci R, Tassinari GA. Neuropathy in with ovarian-cancer. N Engl J Med 1990;322:89–94.
multiple myeloma treated with thalidomide—a prospective study. Neurology [61] von Elm E, Altman DG, Egger M, Pocock SJ, Gotzsche PC, Vandenbroucke JP, for
2007;69:573–81. the STROBE Initiative. The Strengthening the Reporting of Observational
[54] Rao RD, Flynn PJ, Sloan JA, Wong GY, Novotny P, Johnson DB, Gross HM, Renno Studies in Epidemiology (STROBE) statement: guidelines for reporting
SI, Nashawaty M, Loprinzi CL. Efficacy of lamotrigine in the management of observational studies. Lancet 2007;370:1453–7.
chemotherapy-induced peripheral neuropathy—a phase 3 randomized, [62] von Hehn CA, Baron R, Woolf CJ. Deconstructing the neuropathic pain
double-blind, placebo-controlled trial, N01C3. Cancer 2008;112:2802–8. phenotype to reveal neural mechanisms. Neuron 2012;73:638–52.
[55] Rao RD, Michalak JC, Sloan JA, Loprinzi CL, Soori GS, Nikcevich DA, Warner DO, [63] VonSchlippe M, Harland SJ. Cisplatin neurotoxicity in the treatment of
Novotny P, Kutteh LA, Wong GY. Efficacy of gabapentin in the management of metastatic germ cell tumour: time course and prognosis. Br J Cancer
chemotherapy-induced peripheral neuropathy—a phase 3 randomized, 2001;85:823–6.
double-blind, placebo-controlled, crossover trial (NOOC3). Cancer [64] Wallace VCJ, Blackbeard J, Pheby T, Segerdahl AR, Davies M, Hasnie F, Hall S,
2007;110:2110–8. McMahon SB, Rice ASC. Pharmacological, behavioural and mechanistic
[56] Sikandar S, Patel R, Patel S, Sikander S, Bennett DL, Dickenson AH. Genes, analysis of HIV-1 gp120 induced painful neuropathy. PAINÒ 2007;133:47–63.
molecules and patients—emerging topics to guide clinical pain research. Eur J [65] Won HH, Lee J, Park JO, Park YS, Lim HY, Kang WK, Kim JW, Lee SY, Park SH.
Pharmacol 2013;716:188–202. Polymorphic markers associated with severe oxaliplatin-induced, chronic
[57] Smith EML, Pang H, Cirrincione C, Fleishman S, Paskett ED, Ahles T, Bressler LR, peripheral neuropathy in colon cancer patients. Cancer 2012;118:2828–36.
Fadul CE, Knox C, Le-Lindqwister N, Gilman PB, Shapiro CL, for the Alliance for [66] Woolf CJ, Mannion RJ. Neuropathic pain: aetiology, symptoms, mechanisms,
Clinical Trials in Oncology. Effect of duloxetine on pain, function, and quality of and management. Lancet 1999;353:1959–64.
life among patients with chemotherapy-induced painful peripheral [67] Zedan AH, Vilholm OJ. Chemotherapy-induced polyneuropathy: major agents
neuropathy: a randomized clinical trial. JAMA 2013;309:1359–67. and assessment by questionnaires. Basic Clin Pharmacol Toxicol
[58] Sterne JAC, Sutton AJ, Ioannidis JPA, Terrin N, Jones DR, Lau J, Carpenter J, 2014;115:193–200.
Rucker G, Harbord RM, Schmid CH, Tetzlaff J, Deeks JJ, Peters J, Macaskill P,