Dyskeratosis congenita (DC) is an inherited bone marrow failure syndrome (IBMFS) classically diagnosed by the triad of dysplastic nails, reticular skin pigmentation, and oral leukoplakia. Patients with DC are at very high risk of bone marrow failure (BMF), cancer, pulmonary fibrosis, and other medical problems. Germline mutations in key telomere biology genes cause DC (DKC1, TERC, TERT, TINF2, NOP10, NHP2, WRAP53, CTC1 or RTEL1), although about 30% of patients lack a known mutation. Telomere length (TL) below the1st percentile for age in leukocyte subset, measured by flow cytometry with in situhybridization (flow FISH), is diagnostic of DC with a sensitivity and specificity of more than 95%. However, flow FISH requires fresh or cryopreserved blood, and thus is relatively low-throughput and costly. Quantitative PCR (QPCR) of telomeric DNA content, expressed as the ratio of the telomeric product (T) to a single copy gene (S) product, is a high-throughput, inexpensive method of determining relative TL (RTL) that is being used in numerous studies.We evaluated the clinical utility of QPCR RTL measurement as a diagnostic test for DC in the NCI's IRB-approved longitudinal cohort study of IBMFS. Patients were classified as DC (n=31) if they had a mutation in one of the nine known causative genes and/or met the clinical criteria for DC including at least two features of the diagnostic triad with other clinical findings consistent with DC. Fifty-one mutation-negative, healthy relatives of DC patients served as controls. Flow FISH lymphocyte TL was less than or equal to the 1st percentile for age in 29/31 (94%) DC patients and above 1stpercentile in all but one mutation-negative relative. Monoplex QPCR of DNA derived from whole blood was used to determine the RTL (T/S ratio). QPCR RTL data were log transformed and standardized for age using percentile curves and z-scores calculated from 477 healthy age-matched controls. Control samples from children were derived from the control arm of the eating behavior and obesity studies at the National Institute of Child Health and Human Development, NIH, and adult samples were from healthy donors in the National Marrow Donor Program research sample repository. We used Mann-Whitney and Kruskal-Wallis tests to compare z-scores of TL between categories and Receiver Operating Curve (ROC) analysis to calculate the sensitivity and specificity of QPCR RTL at different percentile cut-offs.DC patients had significantly shorter QPCR RTL than relatives (median age adjusted z-score -1.9 vs. 0.09, p<0.0001). Patients with a mutation in TINF2, or DKC1had shorter QPCR relative TL than those with other mutations (p=0.07). DC patients who had at least two of the three features of the diagnostic triad had shorter QPCR relative TL than those who had one or zero features (p=0.05), but there was no difference in QPCR TL between individuals based on severity of bone marrow failure (p=0.2).QPCR RTL was not correlated with flow FISH TL in DC patients (r=0.09, p=0.64). QPCR RTL was only modestly correlated with flow FISH TL in unaffected relatives (r=0.47, p=0.001). QPCR RTL less than the 1st percentile for age missed diagnosing 60% of the known DC patients as having DC (sensitivity=38.7%, and specificity=98%). Only 75% of DC patients had QPCR relative TL below the 10thpercentile for age (sensitivity=74%), as did 12% of the unaffected relatives (specificity=88%).In summary, QPCR RTL measurement less than 1st percentile for age had low sensitivity for identifying patients with DC. Raising the cutoff to the 10th percentile increased the sensitivity, but 12% of healthy individuals were also falsely diagnosed with DC. Our findings raise doubt about using QPCR for clinical diagnosis of DC. This has implications in the interpretation of studies utilizing different TL measurement methods in studies of DC, bone marrow failure, and related telomere biology disorders. With the current limitations of QPCR RTL in identifying DC patients, 6-panel flow FISH TL continues to be the recommended diagnostic test for DC. Disclosures:No relevant conflicts of interest to declare.
Read full abstract