Clinical and Genetic Predictors of Paclitaxel Neurotoxicity Based on Patient-Versus Clinician-Reported Incidence and Severity of Neurotoxicity in the ICON7 Trial

S. B. Park; J. B. Kwok; R. Asher; C. K. Lee; P. Beale; F. Selle; M. Friedlander


Ann Oncol. 2017;28(11):2733-2740. 

In This Article

Abstract and Introduction


Background Chemotherapy-induced peripheral neuropathy (CIPN) is a dose-limiting toxicity of paclitaxel, with no reliable method to identify at-risk patients. We investigated the incidence and risk factors including genetic polymorphisms associated with the development of CIPN based on clinician and patient reporting of neuropathic symptoms.

Patients and methods Risk factors for the development of CIPN were examined in 454 patients treated with paclitaxel/carboplatin from the International Collaboration on Ovarian Neoplasms 7 (ICON7) trial. Neuropathy was graded by clinicians by standard adverse event reporting and by patients utilising OV28 questionnaire. Genetic risk factors were examined by selecting six single nucleotide polymorphisms in genes associated with microtubule function. Risk factors were assessed via dose-to-event cox regression models.

Results Grade >2 neuropathy was reported by clinicians in 28% of patients, while 67% of patients reported 'quite a bit' or 'very much' tingling or numbness. Agreement between clinicians and patients was poor (κ = 0.236, 95% confidence interval, 0.177–0.296, P < 0.001). Older age, bevacizumab treatment and bowel resection were associated with clinician reported CIPN, while older age and volume of residual disease were associated with patient-reported neuropathy. There were no significant associations between clinician-reported neuropathy or patient-reported neuropathy and TUBB2, CEP72 or individual MAPT or GSK3B SNPs, however MAPT additive polymorphisms were associated with patient-reported neuropathy and GSK3B additive polymorphisms were associated with clinician reported CIPN.

Conclusions There was significant discordance between patient- and clinician-reported neurotoxicity. The lack of consensus regarding optimal outcome measures and whose opinion with regard to CIPN takes precedence is a limitation in the investigation of risk factors for CIPN. Care must be taken to select and include patient-reported outcome measures in CIPN assessment to enable accurate identification of genetic and other risk factors for neuropathy.


Paclitaxel (Taxol) results in peripheral nerve damage in up to 70% of ovarian cancer patients, with moderate to severe symptoms in 20%.[1] Chemotherapy-induced peripheral neuropathy (CIPN) impacts on fine motor function and balance, leading to a threefold increase in falls risk.[2] Importantly, CIPN affects physical, functional and emotional well-being and quality of life (QOL) in ovarian cancer patients.[3] This may persist long term, with ovarian cancer survivors reporting neuropathic symptoms 12 years after treatment, which were negatively associated with quality of life.[4]

Clinician-based grading scales are widely used to record neuropathy severity both in routine clinical practice and trials but are limited by inter-observer variability and under-reporting of significant symptoms.[5] Accordingly, patient-reported outcomes (PROs) are increasingly utilised to appreciate the incidence, trajectory and impact of CIPN from the perspective of the patient. However, there remains a lack of consensus regarding appropriate CIPN outcome measures and in particular whose opinion should take precedence.

Accurate assessment of toxicity severity is critical to the successful identification of risk factors associated with CIPN. The lack of reliable and well validated assessment tools has contributed to discordance between studies of clinical and genetic risk factors.[6] There are currently no reliable predictors to identify patients at greater risk of developing persistent neurotoxicity. While a number of genetic polymorphisms have been investigated, contributions to paclitaxel-induced neurotoxicity are likely polygenic.[7] Paclitaxel alters microtubule dynamics and stabilization, which may be important in neuropathy development.[8] Accordingly, single nucleotide polymorphisms (SNPs) in microtubule-associated genes have been associated with paclitaxel[9,10] and vincristine-induced neuropathy.[11]

The International Collaboration on Ovarian Neoplasms 7 (ICON7) trial of paclitaxel/carboplatin chemotherapy with or without bevacizumab included two methods of assessing CIPN, enabling comparison between clinician and patient-reported neuropathy. The aim of this study was to investigate the association between clinician and patient-reported neurotoxicity with demographic and putative genetic risk factors.