Effect of experimental hookworm infection on insulin resistance in people at risk of type 2 diabetes

Abstract

The reduced prevalence of insulin resistance and type 2 diabetes in countries with endemic parasitic worm infections suggests a protective role for worms against metabolic disorders, however clinical evidence has been non-existent. This 2-year randomised, double-blinded clinical trial in Australia of hookworm infection in 40 male and female adults at risk of type 2 diabetes assessed the safety and potential metabolic benefits of treatment with either 20 (n = 14) or 40 (n = 13) Necator americanus larvae (L3) or Placebo (n = 13) (Registration ACTRN12617000818336). Primary outcome was safety defined by adverse events and completion rate. Homoeostatic model assessment of insulin resistance, fasting blood glucose and body mass were key secondary outcomes. Adverse events were more frequent in hookworm-treated participants, where 44% experienced expected gastrointestinal symptoms, but completion rates were comparable to Placebo. Fasting glucose and insulin resistance were lowered in both hookworm-treated groups at 1 year, and body mass was reduced after L3-20 treatment at 2 years. This study suggests hookworm infection is safe in people at risk of type 2 diabetes and associated with improved insulin resistance, warranting further exploration of the benefits of hookworms on metabolic health.

Fig. 1. Study time line, CONSORT flow diagram and study progression.

A After screening and informed consent, participants were allocated to receive either placebo (chilli pepper solution) or a total of 20 (L3-20) or 40 (L3-40) N. americanus hookworm larvae delivered to the skin over 2 occasions, at month 0 and month 2. Participants underwent evaluation visits every 6 months, where physical and vitals exams were undertaken, adverse events were reviewed, and biological samples and questionnaires were collected for analysis of safety, pathology, well-being, diet and exercise habits. B CONSORT chart showing flow of patients through the clinical trial, and reasons for early termination. C Kaplan–Meier analysis of rates of trial continuation (percentage survival and standard error of the mean) in each group during the 2-year study (Placebo n = 13, L3-20 n = 14, L3-40 n = 13; Log-rank test for trend, p = 0.832). Source data are provided as a Source Data File.

Fig. 2. Evaluation of establishment of hookworm infections.

A Peripheral blood eosinophil counts (median ± IQR) at each evaluation visit, shaded area indicates normal range (Sample size for month 0, 6, 12, 18 and 24: Placebo n = 12, 12, 9, 8 and 6; L3-20 n = 13, 12, 11, 11 and 8; L3-40 n = 11, 11, 10, 8 and 8, respectively). # Indicates significantly different to 0 months; Tukey’s posthoc test: (B) Heat map displaying eosinophil counts the 6 month visit (peak eosinophilia) in each participant and correlation with the (C) detection of hookworm eggs in faeces by qPCR (Hookworm patency). Some samples were unable for analysis due to dropping out prior to the 6-month visit. D Quantification of hookworm eggs per gram (EPG) in faecal samples at 6 months (median, and individual data points, L3-20 n = 12, L3-40 n = 9). Source data are provided as a Source Data File.

Fig. 3. Effect of experimental hookworm infection on insulin resistance (HOMA-IR), fasting blood glucose and insulin.

Changes in the (A) Homoeostatic model assessment of insulin resistance (HOMA-IR, units), B fasting blood glucose and (C) fasting blood insulin levels from baseline values at each 6-monthly evaluation visit. Violin plots display each individual data point, median and IQR. Dotted vertical line indicates baseline levels (0 change). *Significant difference to placebo, Kruskal–Wallis test with Dunn’s multiple comparisons. ^#Median value significant difference to baseline, Durbin, Skillings–Mack test and Wilcoxon signed-rank test. Source data are provided as a Source Data File.

Fig. 4. Impact of experimental hookworm infection on body mass.

Changes in participant’s body mass (kg) from baseline values at each 6-monthly evaluation visit. Violin plots display each individual data point, median and IQR. Dotted vertical line indicates baseline levels (0 change). #Median value significant difference to baseline, Durbin, Skillings-Mack test and Wilcoxon signed-rank test: Source data are provided as a Source Data File.