Del Giglio, A.B., Cubero Dde, I., Lerner, T.G., Guariento, R.T., de Azevedo, R.G., Paiva, H., . . . Giglio, A.D. (2013). Purified dry extract of Paullinia cupana (guarana) (PC-18) for chemotherapy-related fatigue in patients with solid tumors: An early discontinuation study. Journal of Dietary Supplements, 10, 325–334. 

To evaluate the effect of dry extract guarana (PC-18) on cancer-related fatigue (CRF) in patients with solid tumors and to evaluate the effect of maintenance doses on CRF in patients who initially improve

PC-18 37.5 mg orally twice daily for 21 days. Those with improved or stable BFI scores were randomized to 37.5 mg BID dosing of PC-18 or placebo for an additional 21 days.

• N = 40 in the induction phase and 33 in the randomization phase
• MEAN AGE = 55.9 years in the induction phase, 52.17 years in the placebo group, and 58 years in the PC-18 group
• MALES: 42.5% induction, 39.4% randomization; FEMALES: 57.5% induction, 60.6% randomization
• KEY DISEASE CHARACTERISTICS: Breast: 25% induction, 30.3% randomization; colorectal: 22.5% induction, 21.2% randomization; lung: 10% induction, 9% randomization; head and neck: 12.5% induction, 9% randomization; other: 30% induction, 30% randomization.
• OTHER KEY SAMPLE CHARACTERISTICS: Cisplatin-based chemotherapy: 25% induction, 24% randomization; doxorubicin-based: 17.5% induction, 15% randomization. All participants receiving FLOX/Xelox/FOLFOX received PC-18 after randomization (15% of randomized participants). All receiving Folfiri also received placebo after randomization (15% of randomized participants).

• SITE: Multi-site  
• SETTING TYPE: Not specified  
• LOCATION: Sao Paulo, Brazil

• PHASE OF CARE: Active anti-tumor treatment

• Stage I, uncontrolled, open study
• Patient showing improvement or stabilized BFI scores were then randomized to continue study drug or placebo.

• Brief Fatigue Inventory
• FACIT-F
• Chalder
• HADS
• PSQI

Mean BFI score decreased by 2.503 points (p = .0002). No significant difference was noted between PC-18 and placebo groups after randomization in BFI (p = .8499), Chalder (p = .6321), FACIT-F (p = .7452), HADS-A (p = .7521), HADS-D (p = .9425), or PSQI (p = .807). There was one instance of grade III depression and one instance of grade III dizziness experienced in PC-18 that was not experienced in placebo. Grade II dizziness and tremors (one instance of each) also reported in PC-18 group, but not in placebo group.

BFI improvement was seen in the induction phase with no significant difference once patients were randomized. It is difficult to make any positive conclusions on guarana as all patients initially had the drug with no washout period before randomization. Potential side effects of guarana may be undesirable.

• Small sample (< 100)
• Risk of bias (no control group)

The study is limited by its sample size and study design. Use of guarana for CRF is not supported by this study.