Scalp cooling for chemotherapy hair loss

Key takeaways

• It works in the right setting. In the 2017 SCALP randomized trial (182 women randomized, 142 analysed at interim), 50.5% of cooled women kept more than half their hair at cycle 4, against 0% of controls (Nangia 2017). A second US cohort using a different system found 66.3% success versus 0% in a small non-randomized control group of 16 (Rugo 2017).
• Pooled success is about 60%. A meta-analysis of 27 studies (2,202 patients) estimated 61% effectiveness with very high heterogeneity (Wang 2021), and a taxane-focused meta-analysis found 60.6% had under 50% hair loss (Lambert 2024).
• Regimen matters. In the 2026 trastuzumab deruxtecan phase II study (40 patients who chose their own arm), cooling did not reduce alopecia: grade 1 or worse in 90% of cooled versus 75% of non-cooled patients (Salehi 2026).
• It may help regrowth afterwards. In a Korean open-label trial of 139 patients, persistent alopecia 6 months after chemotherapy was 13.5% with cooling versus 52.0% without (Kang 2024).
• Longer is not clearly better. Cooling from 30 minutes before to 2 hours after infusion did not beat 30 minutes after infusion in a 37-patient pilot and caused more head discomfort (Carton 2024). For weekly paclitaxel, 20 minutes of post-infusion cooling matched 45 minutes (Lugtenberg 2022).
• Scalp metastases look rare. A pooled analysis reported 0.61% with cooling versus 0.41% without, not significantly different (Rugo 2017 meta-analysis), and a 2025 pooled estimate was 0.15% (Kearney 2025).
• Discomfort is the main cost. Chills (42%), cap heaviness (35%) and headache (30%) were most common, and 18% stopped cooling, mostly because hair loss progressed (Kearney 2025).

What it is

A cap or circulating coolant lowers scalp skin temperature (by about 3 C in the 2017 cohort). The usual explanation is that cold narrows scalp blood vessels so less drug reaches the follicles, and cooled follicle cells take up less drug. For the condition itself see anagen effluvium.

What the studies show

• Automated systems circulate chilled liquid through a fitted cap and monitor temperature. Manual cold caps are frozen gel caps swapped every 20 to 30 minutes. In the 2026 COHAIR trial (152 randomized), hair preservation was 74.5% with an automated system versus 71.2% with a chemical cap, with lower anxiety in the chemical-cap group (Zhang 2026).
• Adding red light on top gave no extra hair retention in a 29-patient trial (Claes 2025).
• Most data are in women with breast cancer, and the two 2017 US studies had manufacturer-affiliated authors. Success definitions differ between studies, and data for blood cancers, very curly hair and high-dose regimens are thin.

How it's used in studies

Studies cooled from about 30 minutes before infusion until 20 to 120 minutes after, depending on the drug. Ask the oncology team about cap fit, especially with dense or tightly coiled hair, where no dedicated trial was found. The US FDA has cleared several automated systems for solid tumours; EU and UK status was not checked.

Side effects and safety

Expect temporary cold intolerance, headache and chills. Scalp metastases appear rare in pooled data, though follow-up is limited.

Bottom line

Scalp cooling is the only widely used way to reduce chemotherapy hair loss, with about 60% success overall and weaker results with some regimens and newer drugs. It guarantees nothing, so raise it with your oncology team before the first cycle. Discuss with a doctor.