We have identified 21 different families with heterozygous coding mutations in the gene encoding the protein component of telomerase, TERT
( and Table S1
). Most of the families have been previously reported
. Eight have been identified more recently and all have been more fully expanded. From our collection of 106 unrelated kindreds with familial pulmonary fibrosis, 19 (18%) have been found with heterozygous TERT
mutations. Previously, we reported that in a group of unrelated individuals with an idiopathic interstitial pneumonia and no family history of pulmonary fibrosis in a first or second-degree family member, 2 (3%) had a TERT
mutation. We have also expanded these two families (S957R and R865C in ) and included them in the analysis. All of the missense mutations involve conserved residues in regions of the protein that have postulated roles in enzyme activity; none of the mutations were reported in a multiethnic panel of 528 controls
or are present in the SNP database. Two of the mutations (V144M and R951W) have been found in different unrelated families
. All the new mutations (R631Q, R671W, V867M, H925Q, R951W and G1063S) have not been identified previously and demonstrate reduced in vitro
telomerase activity as measured by the telomere repeat amplication protocol (TRAP) assay (Figure S1
). All probands have been sequenced for the entire coding exons and flanking introns of TERT
. Only the families with TERT
mutations have been included in this analysis.
Abridged pedigrees of kindreds with heterozygous mutations in the gene encoding the protein component of telomerase (TERT).
Direct genomic DNA sequencing for the proband's TERT mutation in the expanded families identified 134 heterozygous TERT mutation carriers with a mean age of 51 years. For some deceased individuals, archived formalin-fixed paraffin embedded tissue was obtained and directly sequenced for the mutation. Only individuals for whom the mutation was directly sequenced or could be inferred based upon the family structure have been included in our analysis. Most of the mutation carriers come from three large pedigrees (). The self-described ethnicity of 90% of mutation carriers is white, the remainder is Hispanic.
Each subject completed a medical questionnaire including self-reporting of medical diagnoses; these were confirmed with medical records when available. Forty percent of the mutation carriers are affected with pulmonary fibrosis and almost half reported one of the following respiratory diseases: pulmonary fibrosis, obstructive sleep apnea, chronic obstructive pulmonary disease, asthma or previous pneumothorax. Gastroesophageal reflux disease is reported by 26%, with gastritis and/or peptic ulcer disease found in ~8%. Eight percent of patients have elevated liver function tests and for one of these, cryptogenic liver cirrhosis was diagnosed. Osteopenia and/or osteoporosis are reported by over one-fourth of mutation carriers. Over 15% have at least one blood dyscrasia; mild anemia is more common than aplastic anemia or myelodysplastic syndrome combined. The relative frequencies of other diseases are listed in .
Phenotype of 134 subjects with heterozygous TERT mutations.
Telomere length of genomic DNA isolated from circulating leukocytes was determined using a modified multiplexed quantitative PCR assay. In short, this assay assesses the telomere length as a ratio of the telomere copy repeats to a single copy gene, relative to a reference sample. We have observed good correlation between this method and the standard Southern-blot method (terminal restriction fragment length analysis) for determining telomere length (Spearman's rank correlation
0.83, P-value <2.2×10−16
, Figure S2
). We found that those with a TERT
86) have significantly shorter telomere lengths than an unrelated healthy cohort ranging in age from 19–89 years of age (P-value
, ). Approximately 80% of TERT
mutation carriers fell below the 10th
percentile of the reference group; all were shorter than the 50th
percentile (). Most of the TERT
mutation carriers with pulmonary fibrosis, blood dyscrasias or liver cirrhosis were 40 years of age or older. The age-related decline in telomere lengths of the TERT
mutation carriers was similar to that of the normal controls.
Telomere lengths of family members from TERT kindreds.
When comparing telomere lengths between individuals, the values are age-adjusted and reported as an observed minus expected value (O-E value) (). There was no difference between telomere lengths of the reference population and the unrelated spouses, but we find that the mean telomere length of related family members without a TERT
mutation was significantly shorter than the reference population (P-value
). For the three largest kindreds, we were able to estimate a minimal number of generations through which the TERT
mutation segregated. For these three families, the mean telomere length of the TERT
mutation carriers progressively shortened for each TERT
-carrying generation, G3 through G5 () (Linear trend test P-value
0.04). The spread of telomere lengths of TERT
mutation carriers was sufficiently broad to partially overlap the spectrum of lengths seen in the normal controls, spouses and related family members. We also studied the telomere lengths of children of TERT
mutation carriers that did not inherit their grandparents' and parents' mutation (G2+1) and compared these to age-matched controls. Although not statistically significant, there was again a trend toward shorter telomere lengths of children without a TERT
mutation who were born to TERT
carriers corresponding to the number of successive generations the mutation segregated in the family. We investigated the telomere lengths as related by genotype and parental gender (). The telomere lengths of children who inherited a TERT
mutation were shorter if the mutant allele was transmitted from the father rather than the mother, even after adjusting for the sex of the child (P-value
0.01). A TERT
mutation not transmitted from parent to child led to an increase in telomere length, regardless of parental gender, that did not reach the mean telomere length of controls.
Since pulmonary fibrosis is common in TERT mutation carriers, we sought to more carefully characterize the clinical interstitial lung disease phenotype (). We obtained and independently reviewed all available medical records and archived radiographic and pathologic specimens for those with a self-reported or family-reported diagnosis of pulmonary fibrosis. Only older adults were affected, with diagnoses made between 42 to 83 years of age. In general, men were more commonly affected (60%) than women (40%) and had an earlier clinical presentation, with a mean age of 54 vs. 63 years, respectively. Dyspnea and crackles were almost uniformly seen.
Features of 53 TERT mutation carriers with pulmonary fibrosis.
Pulmonary function tests and diffusion capacity measurements were available for a subset of those with pulmonary fibrosis (). All affected individuals had a decrease in the diffusion capacity, a cardinal parameter of IPF. In addition, the majority had evidence of restrictive physiology.
Pulmonary function tests of TERT mutation carriers with pulmonary fibrosis.
Over half were former or current smokers with a mean 21 pack-year cigarette smoking history. Each of the living TERT mutation carriers completed a pulmonary questionnaire that included self-reported drug, radiation, occupational or environmental exposures that have been linked to the development of pulmonary fibrosis. Over ninety-five percent of TERT mutation carriers with pulmonary fibrosis report an exposure to smoking and/or a fibrogenic environmental or occupational agent that may have contributed to the development of their interstitial lung disease. There appears to be a significant association between smoking and/or fibrogenic exposures with pulmonary fibrosis in TERT mutation carriers who are ≥40 years of age. ().
Relationship between smoking, fibrogenic exposures and pulmonary fibrosis in TERT mutation carriers ≥40 years of age.
Radiographs, including CT scans of the chest, were evaluated for 39 different cases. For 29 (74%) subjects, the pattern of pulmonary fibrosis was typical for UIP, that is, there was patchy reticulation concentrated in the periphery and bases which was accompanied by honeycombing (). Honeycombing was generally mild or moderate, but occasionally severe. For 5 subjects, the pattern of fibrosis was consistent with UIP except for an absence of honeycombing. For the remaining 5 subjects, the CT scans were atypical of UIP because the fibrosis was predominantly located in the mid or upper lung fields or along the bronchi. shows representative CT scans typical of UIP, consistent with UIP but without honeycombing, and atypical for UIP. Enlarged mediastinal lymph nodes (exceeding 1 cm in the short axis) were found in 15 (38%) cases. Sixteen cases included expiratory scans, which detected air-trapping in 6 cases. All cases with a radiographic pattern typical or consistent with UIP had a pathologic diagnosis of UIP (n
21; ). Regardless of the specific radiographic pattern, all subjects with serial CT radiographs (n
19) showed progression.
Radiographic and pathologic findings of heterozygous TERT mutation carriers with pulmonary fibrosis.
Clinical diagnosis, radiographic pattern and pathologic features of 29 TERT mutation carriers with surgical lung specimens.
Lung biopsy specimens from 29 cases, including 22 surgical biopsies and 8 explants (both biopsy and explant were available for one case) were reviewed independently. The majority (25 cases or 86%) had diagnostic histologic features of UIP with a characteristic heterogeneous mixture of interstitial fibrosis containing both collagen deposition and fibroblast foci, islands of normal lung, and areas of architectural distortion with parenchymal scarring and/or honeycomb change. In 10 cases (35%) chronic inflammation, consisting of a mixture of lymphocytes and plasma cells, was increased in the scarred areas of the lung and in adjacent interstitium compared to that usually seen in typical UIP. Another unusual feature seen in 5 cases (17%) was the presence of scattered histiocytes and/or small, loose non-necrotizing granulomas within the interstitium (); these are also not usually seen in typical UIP. Areas of acute lung injury, including bronchiolitis obliterans-organizing pneumonia (BOOP) or diffuse alveolar damage (DAD) were superimposed on UIP in 4 cases, and were indicative of the accelerated form of UIP/IPF. Four cases could not be classified as UIP. The diagnosis in one could not be established because of the extent of DAD superimposed on the honeycomb change. Clinically, this subject was diagnosed with interstial pneumonitis and died less than 8 weeks from the start of her symptoms (). One case showed only BOOP along with unclassifiable subpleural fibrosis; this subject died from respiratory failure secondary to “COPD and pulmonary fibrosis” four years from diagnosis. Another case had chronic interstitial pneumonia with fibrosis that could not be further classified; this subject died 14 months after the surgical lung biopsy was obtained. A final case contained only a minute fragment of subpleural scar that was considered insufficient for diagnosis. For all four non-classifiable cases, the biopsies were taken from a single lobe.
Pulmonary fibrosis for TERT
mutation carriers is an age-related phenotype. None of the mutation carriers were diagnosed with pulmonary fibrosis prior to 40 years of age. The penetrance of pulmonary fibrosis increased to 60% and 50% for men and women, respectively, ≥60 years of age (). Individuals heterozygous for TERT
mutations died at an early age. For 29 male mutation carriers, the average age of death was 57.7. For 24 female mutation carriers, the average age of death was 66.6. In comparison, the life expectancy of individuals in the US in 2006 was 75.1 and 80.2 for men and women, respectively
. For most of the TERT
mutation carriers, the cause of death was related to respiratory insufficiency. On average, the mean life expectancy of TERT
mutation carriers with pulmonary fibrosis was 3 years from the time of diagnosis (). We find that a heterozygous TERT
mutation status predicted a clinical outcome of progressive pulmonary fibrosis that mirrors the clinical course of IPF.
Pulmonary fibrosis is a lethal, age-associated phenotype of TERT mutation carriers.