Prevention and Management of Chemotherapy-Induced Peripheral Neuropathy in Survivors of Adult Cancers: ASCO Guideline Update

This is a clinical practice guideline update based on a systematic review of the literature. The panel asked two questions:

1. **Prevention:** Can any agent (drugs, supplements, or other interventions) prevent CIPN from developing in patients receiving neurotoxic chemotherapy?

2. **Treatment:** What works for patients who already have established CIPN?

The interventions tested across the reviewed studies included:

**Prevention trials:** Acetyl-L-carnitine, vitamin E, calcium and magnesium infusions, glutathione, amifostine, omega-3 fatty acids, venlafaxine, gabapentin, lamotrigine, topical amitriptyline/ketamine, exercise, cryotherapy (wearing frozen gloves/socks during chemotherapy), and compression therapy.

**Treatment trials:** Duloxetine, gabapentin, pregabalin, tricyclic antidepressants (nortriptyline, amitriptyline), topical agents (baclofen/amitriptyline/ketamine gel), acupuncture, scrambler therapy, and exercise.

The comparator in most prevention trials was placebo or no intervention. For treatment trials, comparators included placebo, active controls (e.g., duloxetine vs. placebo), or usual care.

Outcome measures included patient-reported neuropathy symptoms (numbness, tingling, pain), clinician-graded neuropathy scales (e.g., National Cancer Institute Common Terminology Criteria for Adverse Events [NCI-CTCAE]), quality of life measures, and functional assessments.

The guideline synthesises data from 45 primary trials and 3 systematic reviews (2 with meta-analyses). The total sample across all studies is not given as a single number, but individual trials ranged from approximately 30 to over 400 participants.

The population was adult cancer survivors (aged 18+) receiving neurotoxic chemotherapy agents including:

Taxanes (paclitaxel, docetaxel)

Platinum compounds (oxaliplatin, cisplatin, carboplatin)

Vinca alkaloids (vincristine)

Bortezomib

Epothilones

Eribulin

Cancer types included breast, colorectal, ovarian, lung, multiple myeloma, and lymphoma. Most patients were receiving chemotherapy with curative or palliative intent. The guideline excludes paediatric patients, radiation-induced neuropathy, and stem-cell transplant-related neuropathy.

The guideline does not mandate a single measurement tool, but the reviewed studies used a mix of:

**Patient-reported outcome measures:** European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire-CIPN20 (EORTC QLQ-CIPN20), which measures sensory, motor, and autonomic symptoms on a 0–100 scale (higher = worse). Also the Functional Assessment of Cancer Therapy/Gynecologic Oncology Group-Neurotoxicity (FACT/GOG-Ntx) scale.

**Clinician-graded scales:** NCI-CTCAE version 4.0 or 5.0, which grades neuropathy from 0 (none) to 4 (life-threatening or disabling).

**Pain scales:** Numeric rating scales (0–10) or visual analogue scales for pain intensity.

**Neurological examination:** Sensory testing (pinprick, light touch, vibration), deep tendon reflexes, and muscle strength.

**Nerve conduction studies:** Electromyography (EMG) and nerve conduction velocity, though these were not routinely used in most trials.

The key distinction: patient-reported outcomes are considered more sensitive and clinically meaningful than clinician-graded scales, because clinicians often under-report neuropathy severity compared to what patients actually experience.

**Design:** This is a clinical practice guideline update based on a targeted systematic literature review. The ASCO Expert Panel (multidisciplinary, including a patient representative) searched PubMed for randomised controlled trials (RCTs) and meta-analyses published between January 1, 2013, and August 28, 2019, with an updated search in February 2020. They used a "signals" approach—designed to identify only new, potentially practice-changing data.

**Inclusion criteria:** RCTs or meta-analyses focusing on chemotherapy-induced neuropathy in cancer survivors, with neuropathy as an important outcome. Excluded: phase I studies, case reports, non-English language, studies with <10 participants, radiation or stem-cell transplant neuropathy.

**Quality assessment:** Systematic reviews were assessed using the AMSTAR tool (scores 8–9 out of 11, indicating good quality). Individual RCTs were assessed for blinding, allocation concealment, sample size, intention-to-treat analysis, and funding sources. Many primary studies had flaws—mainly around lack of blinding, small sample sizes, high dropout rates, and insufficient statistical power.

**What this design can and cannot prove:**

**Can prove:** This is a high-level synthesis of the best available evidence. The systematic review approach with quality assessment provides a reliable summary of what the current literature says. The multidisciplinary panel and public comment period add credibility.

**Cannot prove:** This is not a new experiment. The guideline is only as good as the underlying studies. Many of those studies were small, poorly blinded, or underpowered. The guideline cannot make strong recommendations where high-quality evidence is lacking. It also cannot account for individual patient variation—it gives population-level guidance, not personalised predictions.

**Major methodological weaknesses of the underlying evidence:**

Most prevention trials were negative (no benefit), but many were too small to detect small-to-moderate effects.

Blinding was often inadequate—patients and clinicians knew they were getting an active intervention, which inflates placebo effects.

Heterogeneity across studies: different chemotherapy drugs, different doses, different schedules, different outcome measures. This makes pooling difficult.

Many trials had high dropout rates due to chemotherapy side effects or disease progression, which can bias results.

Industry funding was present in some trials, particularly for drug company-sponsored studies.

**Prevention of CIPN:**

**No agent is recommended for prevention of CIPN.** This reconfirms the 2014 guideline. The evidence does not support any drug, supplement, or intervention to prevent neuropathy from developing.

**Acetyl-L-carnitine should be discouraged.** Two large RCTs (one with 409 patients, one with 269 patients) found that acetyl-L-carnitine actually *worsened* CIPN compared to placebo. In the larger trial, the acetyl-L-carnitine group had significantly more grade 3+ neuropathy (odds ratio not reported, but statistically significant).

**Calcium and magnesium infusions:** Previously thought to prevent oxaliplatin-induced neuropathy, but a large placebo-controlled RCT (N=353) found no benefit. The guideline recommends against their use for prevention.

**Vitamin E:** Mixed results. One small trial showed benefit, but a larger trial found no effect. Not recommended.

**Cryotherapy (frozen gloves/socks):** Several small single-arm studies suggested benefit, but no large RCTs. The guideline says "insufficient evidence" but does not discourage it, as it is low-risk.

**Exercise:** Some evidence that exercise during chemotherapy may reduce neuropathy symptoms, but trials are small and heterogeneous. Not yet recommended as a standard prevention strategy.

**Venlafaxine:** One small RCT (N=50) suggested reduction in acute oxaliplatin-induced neuropathy, but the evidence is too limited for a recommendation.

**Treatment of established CIPN:**

**Duloxetine is the only agent with sufficient evidence to support its use for painful CIPN.** This recommendation is based primarily on one well-designed, double-blind, placebo-controlled crossover RCT (N=231). Patients taking duloxetine (60 mg daily) reported a greater reduction in pain compared to placebo. The mean difference on a 0–10 pain scale was approximately 0.9 points (95% CI, 0.3 to 1.5; p=0.003). This is a modest effect.

**The amount of benefit from duloxetine is limited.** The guideline explicitly states this. A 0.9-point reduction on a 10-point pain scale is clinically meaningful for some patients but not transformative.

**Gabapentin and pregabalin:** No high-quality evidence supports their use for CIPN. One small RCT of gabapentin (N=115) found no benefit over placebo. The guideline recommends against their routine use.

**Tricyclic antidepressants (nortriptyline, amitriptyline):** One small RCT (N=51) of nortriptyline found no benefit. Not recommended.

**Topical agents (baclofen/amitriptyline/ketamine gel):** One small RCT (N=208) found no benefit over placebo. Not recommended.

**Scrambler therapy:** A non-invasive electrical stimulation device. One small RCT (N=50) suggested benefit, but the evidence is insufficient for a recommendation.

**Acupuncture:** One small RCT (N=63) suggested benefit for pain and numbness, but the sham acupuncture group also improved. Insufficient evidence.

**Exercise:** Some evidence that exercise programmes improve function and reduce symptoms in patients with established CIPN, but trials are small.

**Dose modification:**

The guideline recommends that clinicians should assess the appropriateness of dose delaying, dose reduction, substitution, or stopping chemotherapy in patients who develop intolerable neuropathy and/or functional impairment. This is a clinical judgement, not based on specific trial data.

**Duloxetine for painful CIPN:** On average, patients reported a 0.9-point reduction in pain on a 0–10 scale. To put this in context: a 1-point reduction is often considered the minimum clinically important difference for chronic pain. So this is a small but real effect. For every 5–6 patients treated with duloxetine, one would experience a meaningful reduction in pain (number needed to treat ≈ 5–6).

**Acetyl-L-carnitine for prevention:** Not only did it not help, it made neuropathy worse. The magnitude of harm is not precisely quantified, but the effect was large enough to be statistically significant and clinically concerning.

**Cryotherapy:** In single-arm studies, wearing frozen gloves during paclitaxel infusion reduced the incidence of moderate-to-severe neuropathy from ~50% to ~20% in some reports. But without blinded RCTs, these numbers are unreliable (placebo effect, selection bias, regression to the mean).

**Exercise:** Effect sizes are small-to-moderate. One meta-analysis found a standardised mean difference of ~0.3–0.4 for symptom reduction, which is a small effect.

**What the authors acknowledge:**

Many primary studies had flaws in study design, mainly around lack of blinding, small sample sizes, high attrition, and insufficient statistical power.

The included studies were heterogeneous with respect to chemotherapy regimens, outcome measures, and follow-up duration, making direct comparisons difficult.

The guideline is based on a targeted "signals" search, not a comprehensive systematic review of all literature. Some relevant studies may have been missed.

The evidence base for treatment of established CIPN is particularly weak—only one drug (duloxetine) has any support, and that support comes from a single well-designed trial.

**What a critical reader would note:**

**No blinding in most prevention trials:** If patients believe they are getting a protective intervention, they may report fewer symptoms (placebo effect). This is especially problematic for subjective outcomes like pain and tingling.

**Short follow-up:** Many trials only followed patients during chemotherapy or for a few months after. CIPN can persist or worsen for years (the "coasting phenomenon" with oxaliplatin). Long-term outcomes are poorly studied.

**Industry funding:** Some trials were funded by pharmaceutical companies with a financial interest in positive results. This does not invalidate the findings, but it is a known source of bias.

**Publication bias:** Negative trials are less likely to be published. The fact that most prevention trials were negative despite this bias suggests the true effect of most agents is zero or harmful.

**Population limits:** Most trials were in breast and colorectal cancer patients. Results may not generalise to patients receiving other neurotoxic drugs (e.g., bortezomib for multiple myeloma, vincristine for lymphoma).

**No dose-response data:** For duloxetine, only one dose (60 mg daily) was tested. It is unknown whether higher doses would be more effective or whether lower doses would have fewer side effects.

**No head-to-head comparisons:** There are no trials comparing duloxetine to gabapentin, pregabalin, or other agents for CIPN. The recommendation for duloxetine is based on placebo-controlled data only.

**Patient selection:** Many trials excluded patients with pre-existing neuropathy (e.g., from diabetes). This limits generalisability to the real-world population, where many patients have multiple causes of neuropathy.

For someone running their own n=1 experiment (with medical supervision—CIPN is a serious condition that requires oncologist input):

### What to test (specific intervention and dose)

**For prevention (if you are about to start neurotoxic chemotherapy):**

**Cryotherapy:** Wear frozen gloves and socks during each paclitaxel infusion. Keep them on for 15 minutes before, during, and 15 minutes after the infusion. This is low-risk and some evidence suggests benefit, though not proven.

**Exercise:** A structured programme of aerobic exercise (30 minutes, 5x/week) and resistance training (2x/week) during chemotherapy. Start before chemotherapy begins and continue throughout.

**Do NOT test:** Acetyl-L-carnitine (it may worsen neuropathy), calcium/magnesium infusions (no benefit), or high-dose vitamin E (potential toxicity at high doses).

**For treatment (if you already have CIPN):**

**Duloxetine:** Start at 30 mg daily for 1 week, then increase to 60 mg daily. Take with food. Monitor for side effects: nausea, dry mouth, insomnia, sexual dysfunction. Do not stop abruptly (taper over 1–2 weeks).

**Exercise:** Same as above—aerobic and resistance training may reduce symptoms and improve function.

**Topical approaches:** Some patients report benefit from topical compounded gels (e.g., baclofen 10mg/amitriptyline 40mg/ketamine 20mg in a pluronic lecithin organogel), though the trial evidence is negative. A low-risk n=1 test could be: apply to affected areas 3x daily for 2 weeks, track symptoms.

**Scrambler therapy:** If available, a trial of 10–12 sessions over 2 weeks. This is expensive and not widely available.

### Minimum meaningful duration

**Duloxetine:** 4–6 weeks at the target dose (60 mg daily). The crossover trial used 5 weeks per arm.

**Exercise:** At least 8–12 weeks to see changes in neuropathy symptoms. Some benefits may appear earlier (4 weeks).

**Cryotherapy:** Use during every chemotherapy infusion. The effect is acute—you cannot test it outside of the infusion setting.

**Topical agents:** 2–4 weeks of consistent use to assess benefit.

### What to measure (specific metrics)

**Pain intensity:** Use a 0–10 numeric rating scale (0 = no pain, 10 = worst imaginable pain). Record daily at the same time (e.g., evening).

**Numbness/tingling:** Use a 0–10 scale separately for hands and feet. Record weekly.

**Functional impact:** Can you button a shirt? Feel the ground when walking? Pick up small objects? Rate difficulty on a 0–10 scale.

**Quality of life:** Use the EORTC QLQ-CIPN20 (20 questions, free online) or a simple 0–10 "how much does neuropathy bother you?" question.

**Side effects:** For duloxetine, track nausea, sleep quality, mood, and sexual function.

**Chemotherapy dose:** Record whether your oncologist reduces or delays your chemotherapy due to neuropathy. This is a hard outcome.

### Key confounds to control for

**Chemotherapy dose and schedule:** The single biggest predictor