Short telomere length (TL), an indicator of cellular aging and oxidative stress, has been implicated in glucose homeostasis. Additionally, studies have illustrated that the association of TL with health outcomes may vary by age. Yet, data on the association between TL and gestational diabetes mellitus (GDM) are sparse and the potential effect modification by age remains unknown. We prospectively investigated TL in early pregnancy in relation to the subsequent GDM risk in a case-control study of 93 women with GDM and 186 randomly selected controls matched on age, race/ethnicity, and gestational weeks at blood collection. TL was measured using blood samples collected at 10-14 gestational weeks and reported as the T/S ratio, a ratio of telomere repeat length T to copy number of a single copy gene S. Odds ratios (ORs) and 95% confidence intervals (CIs) were estimated using conditional logistic regression adjusted for major risk factors. Overall, TL was not significantly associated with GDM risk. The TL-GDM association was significantly modified by age (Pinteraction = 0.02). Shorter TL in early pregnancy was associated with an increased GDM risk among women <30 years old (adjusted OR comparing the shortest vs. longest tertile: 3.1, 95% CI = 1.2, 8.1), but not associated with GDM risk among women ≥30 years. Our findings suggest that TL in early pregnancy may be implicated in GDM development, particularly among younger women.