Distinguishing Childhood Non-Severe Aplastic Anemia from Dyskeratosis Congenita through Telomere Length Analysis

Telomeres are nucleoprotein complexes at the ends of linear chromosomes with an essential function of protecting the chromosome end. Telomere biology disorders (TBDs) are caused by germline defects that lead to either accelerated telomere shortening or telomere instability, which could impact lots of organs and have variable presentations. In childhood, TBDs typically present with Dyskeratosis Congenita (DC) and have a very high prevalence of bone marrow failure, which causes challenge to differentiation with AA.Though both AA and DC can present with bone marrow failure and pancytopenia, they have totally different treatment and prognosis. And this challenge is particularly pronounced in DC patients lacking typical abnormalities comparing with non-severe aplastic anemia (NSAA) patients.

Because of the expense and accessibility of gene testing, telomere length (TL) measurement is a useful assay for TBDs screening. However, many AA patients exhibit shortened telomeres, posing challenges in differentiation solely based on TL. Previous studies have found that in healthy individuals, TL exhibits differences between adults and children because telomeres shorten faster throughout childhood than in adult life, underscoring the importance of TL characterization in pediatric patients. Our study aims to facilitate differentiation between NSAA and DC based on TL and delineate the TL distribution in pediatric NSAA patients.

We retrospectively collect 62 pediatric NSAA patients and 12 pediatric DC patients under 18 years between 2021 and 2023 in our center. TL was measured using terminal restriction fragments (TRFs) length analysis by Southern blot, which estimates the median telomere length of total leukocytes. We calculated the aTL in terms of the standard deviation (SD) from 90 normal, age-matched, healthy controls (median age, 9.5 years; range, 1.0-18.0 years). The median aTL for NSAA patients was -1.77 (range,-2.68~-0.98) SD and -2.40 (range, -3.01~1.74) SD for DC patients.

We then define short TL (sTL) group as patients with an aTL low than -2.0 SD.We utilize Student's t-test for comparing quantitative data with normal distribution, the Kruskal–Wallis rank-sum test and Fisher's exact test for comparing quantitative data with non-normal distribution. Categorical data is compared using the Chi-square test. Univariate analysis revealed that reticulocyte counts at the time of TL measurement (P=0.051) and disease duration per 12 months (P=0.094) exhibited a trend towards shorter telomeres. While there were no significant association between TL group in terms of patient age, gender, severe bone marrow failure, nor treatment. Multivariate analysis confirmed that disease duration per 12 months (OR 1.307; 95%CI: 1.030-1.712; P=0.036), reticulocyte counts per 10×10⁹/L (OR 1.342; 95%CI: 1.030-1.803; P=0.037) have independent association with short TL in NSAA patients. To determine the optimal cut-off values for reticulocyte counts and disease duration in distinguishing NSAA patients with short TL, ROC analysis was conducted and reticulocyte counts over 60×10⁹/L or disease duration over 36 months were confirmed. We separately compared the aTL in patients whose reticulocyte counts ≥ 60×10⁹/L or disease duration ≥ 36 months with those of DC patients and found no statistically difference, with p-values of 0.901 and 0.358, respectively. Therefore, we suggest that TL measurement was more significant in patients with disease durations less than 36 months or reticulocyte counts under 60×10⁹/L. And when encountering a childhood NSAA patient with reticulocyte counts exceeding 60×10⁹/L or disease duration exceeding 36 months, differentiation from DC based solely on TL becomes challenging. Gene analysis is essential for the differential diagnosis of these patients.

In summary, our study employed the standard method of Southern blot to assess TL changes in children with NSAA and DC. Subsequently, we noted that age-adjusted TL of NSAA patients were higher than DC patients, except for those with reticulocyte counts over 60×10⁹/L or disease duration over 36 months. Reticulocyte counts and duration of the disease at the time of TL measurement emerged as the primary factors associated with short telomeres in children with NSAA.