Circulating Insulin-Like Growth Factor-1 and Risk of Total and 19 Site-Specific Cancers: Cohort Study Analyses from the UK Biobank

Abstract

Background: Insulin-like growth factor-1 (IGF-1) has been implicated in several malignancies, but few studies have examined multiple cancers simultaneously. We sought to conduct systematic assessments of the association between IGF-1 and cancer risk.

Methods: We conducted a prospective analysis between IGF-1 and incident total and 19 site-specific cancers among 412,645 individuals enrolled in the UK Biobank with follow-up to 2016. IGF-1 was measured using blood samples provided at the baseline examination. HR and 95% confidence interval (CI) were calculated with multivariable-adjusted Cox models with IGF-1 modeled both in sex-specific quintiles and continuously.

Results: Participants were followed for a median of 7.2 years. We observed positive associations between circulating IGF-1 and overall cancer risk for both men (HR = 1.03 per 5-nmol/L increment in IGF-1; 95% CI, 1.01-1.06) and women (HR = 1.03; 95% CI, 1.01-1.06). For specific sites, we observed positive associations for breast (HR = 1.10; 95% CI, 1.07-1.14), prostate (1.09; 95% CI, 1.05-1.12), colorectum (1.07; 95% CI, 1.02-1.11), melanoma (1.08; 95% CI, 1.01-1.15), kidney (1.10; 95% CI, 1.00-1.20), and thyroid (1.22; 95% CI, 1.05-1.42) and inverse associations for lung (0.91; 95% CI, 0.86-0.96), ovaries (0.86; 95% CI, 0.77-0.95), head and neck (0.90; 95% CI, 0.82-0.99), and liver (0.32; 95% CI, 0.26-0.38). The inverse association between IGF-1 and lung cancer was observed only in ever-smokers (HR_ever-smoker = 0.88 vs. HR_never-smoker = 1.14; P_interaction = 0.0005). Analyses comparing extreme quintiles were consistent.

Conclusions: IGF-1 is modestly associated with increased risk of total cancer in both men and women but demonstrated divergent associations for site-specific cancers.

Impact: Our study suggests that IGF-1 could serve as a target for cancer prevention or treatment.

Figure 1.

Associations between insulin-like growth factor-1 (per 5-nmol/L increment) and total and site-specific cancers. Associations were adjusted for age (5-year categories), assessment center, the first two principal components, sex (for cancer in both men and women), alcohol consumption (never, special occasions only, 1–3 times per month, 1–2 times per week, 3–4 times per week, daily/almost daily), smoking status (never, former, current), physical activity on the IPAQ scale (low, moderate, high, missing), height in cm (women: <155, 155–159, 160–164, 165–169, ≥170; men: <165, 165–169, 170–174, 175–179, ≥180), BMI in kg/m² (<21, 21–24.9, 25–29.9, 30–34.9, ≥35), waist-to-hip ratio (women: ≤0.80, 0.81–0.85, >0.85; men: ≤0.90, 0.91–0.95, >0.95), and average household income (less than £18,000, £18,000 to £30,999, £31,000 to £51,999, £52,000 to £100,000, greater than £100,000, refused, do not know). For female-specific cancers, we additionally adjusted for menopausal status (premenopausal, postmenopausal, unknown menopausal status), age at menarche (<12, 12–13, ≥14, missing), parity (nulliparous, 1, 2, 3, ≥4), age at first live birth (nulliparous, <20, 20–24, 25–29, ≥30), ever used oral contraceptives (yes, no), ever used hormone replacement therapy (yes, no). For breast, prostate, colorectal, and lung cancer, we additionally adjusted for family history of the respective cancer in a first-degree relative (yes, no).