Background Currently available Interferon-gamma release assays (IGRAs) show a considerable variability in serial testing for latent tuberculosis infection (LTBI). This study offers first results for the new generation IGRA QuantiFERON-TB Gold Plus (QFT-Plus) introduced in 2015 in comparison with its predecessor QuantiFERON-TB Gold In-Tube (QFT-GIT) from serial testing of students with a migration background at German universities. Methods Forty-one students were selected from a previous study. All students with a positive IGRA were asked and 11 agreed to participate in this cohort study. Additionally 30 students with negative IGRA results were selected by chance. Weekly testing with QFT-Plus and QFT-GIT was performed in all individuals over a 4-week period. IGRA variability was evaluated by calculating conversion and reversion rates. Results From 41 participants a total number of 163 serial measurements were analyzed for each IGRA, leading to 122 possible IGRA changes each. QFT-Plus had four conversions and two reversions leading to a conversion rate of 4.3% (4 of 93 possible conversions, 95% CI 1.4–11.3%) and reversion rate of 6.9% (2 of 29 possible reversions, 95% CI 1.2–24.2%). QFT-GIT had 2 conversions and 1 reversion causing slightly lower rates with 2.2% conversions (2 of 91, 95% CI 0.4–8.5%) and 3.2% reversions (1 of 31, 95% CI 0.2–18.5%). Inconsistent IGRA results occurred in 4 subjects for QFT-Plus (8 stable positives, 29 stable negatives) and in 2 subjects for QFT-GIT (9 stable positives, 30 stable negatives). Agreement between the two IGRAs was 95.1% (κ = 0.89). Variance attributed to the individuals was low (QFT-Plus: ICC = 0.88). Conclusion This study confirms occurrence of conversions and reversions for the new QFT-Plus in serial testing of a high-risk cohort in a low-incidence setting with improbable new TB contact during the study. QFT-Plus conversion and reversion rates were slightly higher than for the QFT-GIT but overall they were lower for both IGRAs than in other studies that investigated IGRA variability. Electronic supplementary material The online version of this article (doi:10.1186/s12995-016-0148-z) contains supplementary material, which is available to authorized users.