Prospective and longitudinal evaluations of telomere length of circulating DNA as a risk predictor of hepatocellular carcinoma in HBV patients
Prospective and longitudinal evaluations of telomere length of circulating DNA as a risk predictor of hepatocellular carcinoma in HBV patients
Shaogui Wan 1 2
Hie-Won Hann 0 5
Zhong Ye 2
Richard S.Hann 0
Yinzhi Lai 2
Chun Wang 2
Ling Li 2
Ronald E.Myers 2
Bingshan Li 4
Jinliang Xing 3
Hushan Yang 2
0 Department of Medicine, Liver Disease Prevention Center, Thomas Jefferson University , Philadelphia, PA 19107 , USA
1 ,Institute of Pharmacy, Pharmaceutical College, Henan University , Kaifeng, Henan 475004 , China
2 Division of Population Science, Department of Medical Oncology, Sidney Kimmel Cancer Center, Thomas Jefferson University , Philadelphia, PA 19107 , USA
3 dState Key Laboratory of Cancer Biology and Experimental Teaching Center, College of Basic Medicine, Fourth Military Medical University , Xi'an 710032 , China
4 Department of Molecular Physiology and Biophysics, Center for Human Genetics Research, Vanderbilt University , Nashville, TN 37232 , USA an
5 Division of Gastroenterology and Hepatology, Department of Medicine, Thomas Jefferson University , Philadelphia, PA 19107 , USA
Prospective and longitudinal epidemiological evidence is needed to assess the association between telomere length and risk of hepatocellular carcinoma (HCC). In 323 cancer-free Korean-American HBV patients with 1-year exclusion window (followed for >1 year and did not develop HCC within 1 year), we measured the relative telomere length (RTL) in baseline serum DNAs and conducted extensive prospective and longitudinal analyses to assess RTL-HCC relationship. We found that long baseline RTL conferred an increased HCC risk compared to short RTL [hazard ratio (HR) = 4P. 9=3 0,.0005). The association remained prominent when the analysis was restricted to patients with a more stringent 5-year exclusion window (HR = 7.51, P = 0.012), indicating that the association was unlikely due to including undetected HCC patients in the cohort, thus minimizing the reverse-causation limitation in most retrospective studies. Adding baseline RTL to demographic variables increased the discrimination accuracy of the time-dependent receiver operating characteristic analysis from 0.769 to 0.868 P( = 1.0 × 10−5). In a nested longitudinal subcohort of 16 matched cases-control pairs, using a mixed effects model, we observed a trend of increased RTL in cases and decreased RTL in controls along 5 years of followup, with a significant interaction of case/control status with tiPmfeor( interaction=0.002) and confirmed the association between long RTL and HCC risk [odds ratio [OR] = 3.63P, = 0.016]. In summary, serum DNA RTL may be a novel non-invasive prospective marker of HBV-related HCC. Independent studies are necessary to validate and generalize this finding in diverse populations and assess the clinical applicability of RTL in HCC prediction.
Telomeres consist of small tandem nucleotide repeats (TTAGGG protect the linear chromosome ends1(). In normal somatic cells,
in humans) that form the physical ends of eukaryotic chrom-o telomeres shorten by 50–200 base pairs within each cell division
somes ( 1). The main functions of telomeres are to stabilize and due to end replication inefficiency of DNA polymerase2(). The
Abbreviations medical chart review and/or consultation with the treating physicians.
ALT alanine transaminase The details on data collection and patient diagnosis were described pr-evi
AFP alpha fetoprotein ously (
). Liver cirrhosis was diagnosed mainly through imaging stu-d
AST aspartate transaminase ies, which showed medial segment atrophy of the left lobe, caudate lobe
AUC area under the curve hypertrophy, or liver nodularity for early stage disease, and signs of portal
hypertension such as varices, splenomegaly, patent paraumbilical vein or
CV coefficient of variation ascites in advanced stage disease. Other criteria were used to complement
HCC hepatocellular carcinoma to imaging studies, including liver biopsy, laboratory tests such as thr-om
HCV hepatitis C viral bocytopenia, serum albumin and prolonged prothrombin time and clinical
HBV hepatitis B virus presentations such as ascites, encephalopathy and gastrointestinal bl-eed
HR hazard ratio ing, etc. The majority of patients have available serum samples collected
RTL relative telomere length at initial study entry as well as subsequent follow-up visits. Among more
ROC receiver operating characteristic than 2600 patients in this cohort, >90% were of Korean ancestry and had
chronic HBV infection. To minimize the confounding of patient ethnicity
and disease etiology, we restricted this study to Korean HBV patients. The
homeostasis of telomeres is regulated by telomerase. The re-la patients included in the primary prospective cohort also met the foll-ow
tionship between telomerase activity, telomere length main- te ing two additional criteria: (i) had a 1-year exclusion window (followed
nance and tumorigenesis is complex and remains a hot research for >1 year and did not develop HCC within 1 year); and (ii) had available
topic (3–5). major demographic data including age, gender, smoking status, drinking
Retrospective case–control studies have shaped the p-er status, family history of cancer and cirrhosis status. For the pilot lon-gitu
dinal case–control analysis, from the 37 patients that were cancer-free at
ception that short relative telomere length (RTL) confers an the time of initial sample collection but subsequently developed HCC, we
increased cancer risk; however, this perception has been selected 16 who had at least two serum samples available within 5 years
increasingly challenged in recent studies that used prospective of follow-up before HCC diagnosis and the first sample was collected
approaches (6–15). Although telomere dysfunction has been s-ig >2 years before HCC diagnosis. Sixteen controls were selected from 286
nificantly implicated in hepatocarcinogenesis in various basic HBV patients who remained cancer-free at their last follow-up and met
), to date, there has not been a report prospe-c the same criteria as the cases. Controls were frequency-matched to cases
tively and longitudinally evaluating RTL and liver cancer risk inon age of first sample collection, age of last sample collection, gender,
hepatitis B virus (HBV) patients. A recent epidemiological study smoking status, drinking status, family history of cancer, cirrhosis status
reported an association of long telomeres with increased risk at first sample collection and follow-up time. This study was approved by
the Institutional Review Board of Thomas Jefferson University. A written
informed consent was obtained from each patient.
of hepatocellular carcinoma (HCC)2(0). Consistently, we found
that longer telomere length in circulating cell-free serum DNA
was significantly associated with an increased risk of both-cir DNA isolation and RTL measurement
rhosis and HCC (
). Nonetheless, all these studies were all Circulating cell-free serum DNA was isolated from 200 ul serum sample
based on a retrospective case–control design and thus were using QIAamp DNA Blood Mini kit (Qiagen, Hilden, Germany) according
limited by the reverse-causation issue inherent in case–control to the manufacturer’s protocol. The RTL of each DNA sample was det-er
studies. Interestingly, two recent meta-analyses indicated that mined by quantitative real-time polymerase chain reaction (qRT-PCR),
the significant associations between telomere length and -can which measured the ratio of the copy number of telomere repeats to the
cer risk were observed mostly in retrospective but not the a few copy number of a human single copy gene (36B4). The detailed procedure
prospective studies (
). Taken together, these contradictory of RTL measurement was described previously (
), with the following
findings highlight the importance of prospective evaluations minor modifications in the present study: if the quantity of single RTL or
related to the role of telomere length in cancer risk assessment. single copy gene 36B4 was out of the acceptable range of the standard
Another challenge associated with using telomere length curve, or the cycle threshold (Ct) was >35, the sample was repeated. If
the Ct value of RTL was >35 but the Ct value of single copy ge3n6eB4 was
as a surrogate cancer biomarker is to assessing the dynamic ≤35 and within the acceptable range of the standard curve, the sample
changes of telomere length over time in blood cell2s5–(27). was assigned to the category of short RTL when analyzed as a categorical
Several recent longitudinal studies suggested cross-sectionalvariable, or a small value of 0.0001 when analyzed as a continuous v-ari
measurements limited the analyses of blood cell telomere at-tri able. Dot blot assay was used to assess the specificity of the telomere DNA
tion or change rate that are affected by a variety of factors likemeasured by qRT-PCR and was conducted as previously described3(0). We
age of individuals and environmental stimuli25(–29). As yet, quantified telomere length of one sample in hexaplicate in a single run,
no longitudinal study has been reported to assess the effect of and determined that the intra-assay coefficient of variation (CV) for our
time-dependent telomere length change on cancer devel-op RTL analysis was 2.58%. We also repeated the experiments in duplicate
ment. Using a retrospective approach, we recently reported the samples in four independent runs and determined that the inter-assay
associations between circulating RTL and the risk of cirrhosis aCVsimwialsar4.m67e%t h.Tohde(3s1e).data were comparable to the reports of others using
and HCC (
). In the present study, we conducted a prospe-c
tive cohort analysis and a pilot longitudinal case–control a-naly Statistical analysis
sis to further elucidate the role of telomere length in predicting
HCC risk in patients with chronic HBV infection.
Statistical analyses were performed using the SAS software version 9.3
(SAS Institute, Cary, NC) and Stata 12.0 (StataCorp, College Station, TX). To
include the largest possible sample size, we performed the primary pr-o
Materials and methods spective analysis with a 1-year exclusion window. To minimize the co-n
founding effects of those patients who actually had HCC, but were not
Study population diagnosed at initial sample collection, we further restricted the analyses
Subjects in this study were identified from an existing clinic-based cohort. to subcohorts of patients with a 2-year or 5-year exclusion window. The
Patient enrollment in this cohort started from 1988 and is still ongoing. distributions of host variables and HCC risk were analyzed by chi-square
Patients were those who visited the Liver Disease Prevention Center at test. In prospective analyses, the RTL–HCC association was estimated as
Thomas Jefferson University Hospital for treatment of various liver -dis hazards ratio (HR) and 95% confidence interval (95% CI) by Cox prop-or
eases, such as chronic HBV or HCV infection, fibrosis, cirrhosis or HCC. tional hazards regression model, using a univariate analysis and a mu-lti
Demographic and clinical data were obtained for each patient through variate analysis adjusting for age, gender, smoking status, drinking status,
family history of cancer and cirrhosis status, where appropriate. Dose- the primary prospective cohort of patients with a 1-year exclusion
dependent effect was analyzed using fractional polynomial regression window, dichotomization analyses showed a significant assoc-ia
model adjusting for all the host variable3s2)(. The cumulative incidence tion between longer RTL and HCC risk in both univariate analysis
of HCC by follow-up years was derived using the Nelson–Aalen method (HR = 6.43, 95% CI 2.68–15.43, P = 3.1 × 10−5) and multivariate anal-y
(a3f3t)e.rDiinsictriiamlinsaamtipolne caoclcluercatciyonfoursinprgedRiTcLtinangdH/oCrCmraijsokr whoitshtinch1a0r aycet-aerrsis sis (HR = 4.93, 95% CI 2.00–12.13, P = 0.0005). In the univariate te-r
tics was assessed by the area under the curve (AUC) of time-dependent tile analysis, we observed a significant dose-dependent increase
receiver operating characteristic (ROC) curves for censored survival data in HCC risk along with increasing baseline RTL. Using RTL value
(34). The differences in discrimination accuracy between different ROC in the first tertile as reference, RTL in the second and third- ter
models were assessed by an internal validation using 10 000 bootstrap tile was associated with an HCC risk of 3.31 (95% CI 0.92–11.91,
resampling. In the pilot longitudinal analysis, to analyze the relationship P = 0.067) and 10.09 (95% CI 3.03–33.69, P = 0.0002), respectively,
between the dynamic RTL change and HCC risk, RTL values measured (P for trend=8.3× 10−6) (Table 2). Multivariate analysis yielded very
within 5 years after initial sample collection for 16 matched case–control similar resultsT(able 2). A similar result was also noted when RTL
pairs were plotted against follow-up time. Mixed effects model was used was analyzed as a continuous variable using fractional polynomial
to analyze RTL data through an interaction of the case/control status withregression model (P value of 4.5 × 10−14) (Figure 1). However, the
time, and test the hypothesis that HCC risk is a function of longitudinal trend appeared to be unstable, as reflected by a wide confidence
RTL change by estimating odds ratio (OR) and 95% CI. All statistical tests in
this study were two-sided, and aP value of <0.05 was considered statis-ti
interval, in the few patients with the highest RTLs, likely due to
the small patient number in this RTL rangeFi(gure 1). We further
restricted the prospective analyses to patients with a 2-year (258
patients) or 5-year (153 patients) exclusion window. The longer
Results exclusion window helped minimize the confounding effect of
undiagnosed HCC patients at the time of baseline sample coll-ec
Characteristics of the study population tion. The results of analyses with longer exclusion windows were
There were 344 cancer-free HBV patients who met the criteria highly consistent with that of the analysis with a 1-year ex-clu
of 1-year exclusion window as described earlier. After excluding sion window (Table 2), either in the dichotomization or the tertile
21 patients that failed RTL measurement, 323 patients with a analysis. For instance, in the analysis with a 5-year exclusion w-in
median age of 43.9 (interquartile 25–75% range, 37.6–51.7) were dow, only 2 of the 77 patients with a short baseline RTL developed
included in the primary prospective cohort. During a median HCC but 13 of the 76 patients with a long RTL developed HCC after
follow-up of 4.5 (interquartile 25–75% range, 2.4–8.0) years, 37 5 years (HR = 7.51, 95% CI 1.56–36.18, P = 0.012). Consistent with
of the 323 HBV patients developed HCC (incidence rate, 11.5%). the Cox analyses, cumulative incidences of HCC assessed by the
Detailed distributions of host characteristics of these patients Nelson-Aalen method were significantly higher in patients with
are summarized in Table 1. longer than shorter RTLFi(gure 2). Taken together, these lines of
evidence suggested that RTL of baseline serum DNA might be an
The prospective association of RTL with HCC risk in independent prospective HCC predictor in HBV patients. To assess
HBV patients the specificity of the telomeric DNA measurement in our serum
We analyzed the association between baseline RTL and HCC risk samples, we conducted dot blot analysis to quantify the telomere
using univariate and multivariate Cox model by categorizing RTL content in all the 323 samples tested in this study according to a
into two levels (dichotomization analysis using a cut-off of the published method (
), and analyzed the correlation between the
median RTL value in all subjects) and three levels (tertile an-aly results of dot blot and qRT-PCR methods. We found that dot blot
sis using cut-offs of tertile RTL values in all subjectTsa)b(le 2). In experiment had lower sensitivity than qRT-PCR. Twelve samples
Number of HCC
discrimination accuracy F(igure 3). In the primary prospective
cohort with a 1-year exclusion window, the AUC was 0.678,
0.802 and 0.837 for models with RTL only, demographic var-i
ables (age, gender, smoking status, drinking status, family h-is
tory of cancer and cirrhosis) only (Demo model) and RTL plus
demographic variables (Full model), respectively. The increase
of AUC from Demo to full model was statistically significant
(P = 0.0126) by bootstrap analysis F(igure 3A). Very similar
findings were observed when the analyses were done in
subcohorts with a 2-yearF(igure 3B) or 5-year (Figure 3C)
exclusion window. An interesting observation is that the AUC of
the RTL only model increased from 0.678 in the analysis with
a 1-year exclusion window to 0.710 and 0.773 in the analysis
with a 2-year and 5-year exclusion window, respectively. The
progressive increase of AUC of the RTL only model in the ana-ly
Figure 1. Dose-dependent effect of RTL on the risk of HBV-related HCC. Fr-ac sis with longer exclusion windows further suggest a prosp-ec
tional polynomial regression model was used to analyze the association tive predictive role of baseline RTL in the risk of HBV-related
between baseline RTL and HCC risk. Solid line represents hazards ratio in log HCC. In addition, we compared the performance of RTL to
scale; shaded area indicates 95% confidence interval. Solid dots represent HRs each of the three important clinical markers, including alanine
for each individual. transaminase (ALT), aspartate transaminase (AST) and
alphafetoprotein (AFP) 3(5) in our population. We found the AUC with
were labeled by ImageQuant as undetectable (no signal) and 151 a 1-year, 2-year and 5-year exclusion window was 0.578, 0.575
as having high background noise. Among samples with a detected and 0.563 for ALT, respectively, 0.609, 0.603 and 0.595 for AST,
dot blot signal, we observed a high correlation between the two respectively, and 0.728, 0.731 and 0.743 for AFP, respectively
methods (r = −0.658, P = 6.4 × 10−40) (Supplementary Figure 1, avail- (Supplementary Figure 2, available atCarcinogenesis Online).
able atCarcinogenesis Online). A base model adding these three variables to the six dem-o
graphic variables in the Demo model ofFigure 3 resulted in
HCC risk prediction models incorporating an AUC of 0.824, which increased to 0.871 when RTL was fu-r
baseline RTL ther added (P = 0.029) (Supplementary Figure 3A, available
We constructed 10-year time-dependent ROC curves and at Carcinogenesis Online). When HBV DNA load, another co-n
calculated the AUC in different models to evaluate their sistently reported HCC marker3(6) was further added, the
cases and controls had cirrhosis at sample collectioPn =( 1.00)
(Figure 4A). Using mixed effects model, we observed a trend
of increased RTL in cases and decreased RTL in controls along
5 years of follow-up, with a significant interaction of case/c-on
trol status with timeP(for interaction, 0.002)F(igure 4B). We also
found that, compared to patients with a longitudinal trend of
decreased RTL, those with a longitudinal trend of RTL increase
had a significantly higher HCC risk in both univariate (OR = 3.88,
95% CI 1.32–11.39, P = 0.014) and multivariate (OR = 3.63, 95% CI
1.27–10.35, P = 0.016) analysis (Figure 4C).
In this study, we used prospective and longitudinal approaches
to evaluate the predictive role of serum DNA RTL in HCC risk in a
clinic-based HBV patient cohort. Our results indicated that long
baseline RTL conferred an elevated HCC risk in a dose-depen-d
ent manner.The RTL-HCC association remains highly significant
when the analysis was conducted with a strict 5-year exclusion
window which largely eliminated the confounding effects of
undiagnosed early-stage HCC patients. We also found that a-dd
ing baseline RTL to common demographic and clinical variables
increased the power of predicting HCC risk. Collectively, our
data suggest that serum DNA RTL may be a potential novel
noninvasive HCC risk predictor in HBV patients.
There is a large body of evidence suggesting that shortened
telomeres may play a causal role in cancer development by
inducing chromosomal instability and promoting neoplastic
transformation 3(8). However, an increasing number of recent
studies reported that abnormal telomere length, either sh-ort
ened or lengthened, was linked to various cancers23(,39).
Nonetheless, most of these studies adopted a retrospective
case–control design using samples collected at the time after
cancer diagnosis. Thus, their conclusions were constrained by
the reverse causation limitation23(). Indeed, several more recent
prospective studies failed to replicate the RTL-cancer asso-cia
tions reported in previous case–control studies in the same po-p
). These controversial findings highlight the need
of prospective and longitudinal evaluations of the causal r-ela
Figure 2. Cumulative incidence of HBV-related HCC. The cumulative incidence tionship between RTL and cancer risk. We conducted an exte-n
of HCC risk was derived using the Nelson-Aalen method by dichotomizing RTL sive literature search and identified 25 studies that prospectively
based on a cut-off of median RTL, conducted in patients with a 1-yearA)(, 2-year evaluated the associations between telomere length and cancer
(B) or 5-year (C) exclusion window. risk (Supplementary Table 1, available aCtarcinogenesis Online).
Among which, 12 studies reported an association of increased
AUC was 0.849 for the base model and 0.893 when RTL was cancer risk with long telomeres and 4 studies reported the op-po
included (P = 0.0009) (Supplementary Figure 3B, available at site. Two studies reported aU-shape association indicating that
Carcinogenesis Online). both very short and very long telomeres were associated with an
increased cancer risk. The rest studies did not observe a sign-ifi
Longitudinal analysis cant association. Of these studies, an elegant one (with the l-arg
Several recent longitudinal studies reported a feature of dynamic est population size and longest follow-up) was reported recently
change in telomere length of peripheral blood leukocyt2e5s–(29). by Weischer et al. (9) that prospectively followed 47 102 Danish
The concentration and components of circulating cell-free DNA general population participants for up to 20 years and eva-lu
vary in a wide range that is affected by physiological and pa-th ated the associations between telomere length and 24 major
ological status of individuals37(). Therefore, a single baseline cancer types. They found the risks of breast cancer and sarcoma
RTL measurement may not be reliable enough to yield robust increased with long telomere length but there was no sig-nifi
findings. To help address this concern, we conducted a pilot cant association for the rest 22 cancers. Taking these lines of-evi
longitudinal case–control analysis by measuring all the a-vail dence together, prospective studies seemed to suggest that the
able serum samples collected during 5 years after baseline RTL RTL-cancer risk associations are in a cancer-type specific ma-n
measurement before HCC diagnosis, from 16 matched case–con- ner and each cancer should be analyzed separately. However, it
trol pairs. The cases and controls were adequately matched on is worth noting that the Danish study did not observe a sig-nifi
age at first sample collectionP (= 0.66), age at last sample colle-c cant association between RTL and liver cancer risk either, which
tion (P = 0.80), follow-up time (P = 0.10), gender (P = 0.37), smok - is inconsistent with our study and might be due to the different
ing status P( = 0.72), drinking statusP( = 0.26), family history of population characteristics. For instance, the Danish study was
cancer (P = 1.00) and number of RTL measurements (P = 0.12). All based on a population-based cohort of general healthy white
participants with low HBV infection and low liver cancer i-nci of circulating DNA telomeres in HCC risk. However, the mech-a
dence (0.15%) whereas our study was based on a clinic-based nism underlying the paradoxical observations between blood
cohort of Asian American HBV patients with high HCC incidence and tissue telomere length in the development of HCC in HBV
(11.5%). The liver cancer etiologies are likely to be different, patients, as well as whether our conclusion can be generalized
because unlike other common causes of HCC such as hepat-i to other ethnic groups (non-Asian ancestry), or HCC etiologies
tis C viral (HCV) infection and chronic alcoholic liver diseases (HCV-HCC, alcoholic HCC, etc.) remains to be further evaluated.
that are primarily mediated by progression through liver cirr-ho In the pilot longitudinal analysis, we observed a decreasing
sis, a major mechanism by which HCC may arise from chronic trend of RTL along with age in HBV patients who did not develop
HBV infection is the integration of HBV genome into the host HCC, which was not surprising because a negative correlation
genome, resulting in genomic aberrations that lead to oncogene between telomere length and age has been well-documented. In
activation, tumor-suppressor gene inactivation, or other pre-dis comparison, a significant trend of increasing RTL was noticed in
positions to chromosomal instability4(3). In two recent case– patients who developed HCC after 5 years of follow-upFi(gure 4).
control studies, we reported that long telomeres in circulatingThis data echo the findings of the prospective analysis, f-ur
DNAs were significantly associated with increased risks of -cir ther establishing a temporal link between RTL and HCC risk.
rhosis and HCC in HBV patients, which was consistent with the However, the result of this analysis needs to be interpreted with
findings of another case–control study that assessed RTL in DNA caution due to its unplanned nature and small size.
of peripheral blood leukocytes2(
). Our finding in the present This study has several strengths. The unique and homo-g
study took a step further to confirm a prospective predictive role enous Korean American HBV patient cohort enrolled in a single
institute is a major strength of this study, which eliminated the human samples ( 51). These directions are important to future
confounding effects of patient ethnicity and disease etiology. in-depth mechanistic investigations of telomere length al-ter
Another strength is the restriction of the prospective analyses toation in HCC development.
sub-cohorts with an exclusion window of as long as 5 years that In summary, our study suggests that telomere length in
largely minimized the confounding resulting from the possible circulating cell-free serum DNA may potentially be used as a
inclusion of undetected HCC cases. Moreover, the non-invasive prospective non-invasive marker of HCC risk in HBV patients.
nature of measuring circulating cell-free DNA derived from r-ou Independent populations are necessary to validate and gener-al
tine blood tests makes it possible for repetitive monitoring of ize this finding in diverse populations and assess the clinical
the dynamic change of telomere length during patient follow-up applicability of RTL in HCC prediction.
Our study also has apparent limitations. First, although our Funding
study benefited from a unique prospective and longitudinal
design, our number of HCC patients is modest and the fin-d The work was supported by a Tobacco Grant from the
ings still warrant independent validations. Second, there is a Pennsylvania Department of Health, National Cancer Institute
large body of basic studies linking shortened telomere length Grants (CA153099 and CA159047), American Cancer Society
and tumorigenesis. However, there is very scarce experime-n Research Scholar Grant (123741-RSG-13-003-01-CCE) and
tal evidence supporting a link between elongated telomere Institutional Research Grant (0806001) and a Research Scholar
length and cancer development. Although this may be pa-r Award from the V Foundation for Cancer Research.
tially accounted for by different characteristics and functionsConflict of Interest Statement: None declared.
between tissue and blood cells19(,44–46), mechanistic studies
are needed to elucidate the molecular mechanisms under-ly Supplementary material
ing the associations between long RTL and elevated cancer
risk observed in various epidemiological studies. Third, recent Supplementary data are available aCtarcinogenesis online.
studies reported that demographic and clinical variables
such as obesity, glucose intolerance, des-gamma carbox-y
prothrombin, transaminases and AFP, were associated with References
HCC risk (35,47–49). Since the data of our study were obtained 1. Blackburn, E.H. (1991) Telomeres. Trends Biochem. Sci., 16, 378–381.
through medical chart review, these variables were not av-ail 2. Klapper, W. et al. (2001) Telomere biology in human aging and aging
able because they were either not completely recorded, or syndromes. Mech. Ageing Dev., 122, 695–712.
not measured at the time of blood collections. It is also worth 3. Harley, C.B.et al. (1990) Telomeres shorten during ageing of human
noting that, compared to the study of Wenet al. (35). which 4. fFiibnrkoebll,aTs.etts a.lN.(a2t0u07r)eT,3h4e5c,o4m58m–o4n60b.iology of cancer and ageing. Nature,
reported an AUC of >0.9 in a model that included age, gender, 448, 767–774.
ALT and AST, our current study reported much lower AUCs on 5. Shay, J.W. (2016) Role of telomeres and telomerase in aging and cancer.
models based on ALT, AST or AFP individually (Supplementary Cancer Discov., 6, 584–593.
Figure 2, available atCarcinogenesis Online). The differences in 6. Lynch, S.M. et al. (2013) A prospective analysis of telomere length and
predictive powers were likely due to the differences between pancreatic cancer in the alpha-tocopherol beta-carotene cancer (ATBC)
the two studies in population characteristics, sample sizes, prevention study. Int. J. Cancer, 133, 2672–2680.
variables that were included in the models, and cutoffs used 7. Nan, H. et al. (2011) Shorter telomeres associate with a reduced risk of
for the variables. Therefore, it remains to be further eva-lu melanoma development. Cancer Res., 71, 6758–6763.
ated if RTL adds additional predictive value to a model that 8. Lan, Q. et al. (2009) A prospective study of telomere length measured
by monochrome multiplex quantitative PCR and risk of non-Hodgkin
includes a more comprehensive panel of variables. Fourth, c-ir lymphoma. Clin. Cancer Res., 15, 7429–7433.
culating cell-free DNA is a heterogeneous mixture including 9. Weischer, M. et al. (2013) Short telomere length, cancer survival, and
DNA molecules from a wide spectrum of cell types with broad cancer risk in 47102 individuals. J. Natl. Cancer Inst., 105, 459–468.
variations in content and concentration5s0)(. Whether the 10. Han, J. et al. (2009) A prospective study of telomere length and the risk
telomere DNA measured in this study is from a combination of skin cancer. J. Invest. Dermatol., 129, 415–421.
of all or some of these cell types is an important question w-or 11. Hosnijeh, F.S. et al. (2014) Prediagnostic telomere length and risk of
thy of further investigations. In our current study, because the B-cell lymphoma-Results from the EPIC cohort study. Int. J. Cancer, 135,
majority of the patients are either cancer-free HBV patients 2910–2917.
or HCC patients with early-stage tumors, it is reasonable to 12. Seow, W.J. et al. (2014) Telomere length in white blood cell DNA and lung
conjecture that the circulating DNAs of these patients should c40a9n0c–e4r0:9a8p.ooled analysis of three prospective cohorts. Cancer Res., 74,
mainly be derived from normal blood DNAs, thus reflecting 13. Machiela, M.J.et al. (2015) Genetic variants associated with longer -tel
the genetic background of the subjects. However, normal or omere length are associated with increased lung cancer risk among
tumor hepatocytes may also significantly contribute to ci-rcu never-smoking women in Asia: a report from the female lung cancer
lating DNAs because in HBV patients, chronic inflammation consortium in Asia. Int. J. Cancer, 137, 311–319.
tends to lead to repeated cycles of destruction and regen-era 14. Julin, B. et al. (2015) Circulating leukocyte telomere length and risk of
tion of liver tissues, a process that releases a large amount overall and aggressive prostate cancer. Br. J. Cancer, 112, 769–776.
of apoptotic or necrotic hepatocytes into circulation. Thus, to 15. Machiela, M.J.et al. (2016) Genetically predicted longer telomere length
more accurately determine the cell type origin of circulating is associated with increased risk of B-cell lymphoma subtypes. Hum.
DNA RTLs, it is important to compare RTLs between diffe-r 16. KMioml,.HG.enete ta.l,. 2(250,0196)6L3a–1r6g7e6l.iver cell change in hepatitis B virus-related
ent cell types such as serum, leukocytes, tumor tissue and liver cirrhosis. Hepatology, 50, 752–762.
normal tissue that are obtained at the same time from the 17. Lee, Y.H. et al. (2009) Chromosomal instability, telomere shortening,
same patients. Another possible solution is to use high-depth and inactivation of p21(WAF1/CIP1) in dysplastic nodules of hepatitis
next-generation sequencing technology to sequence the t-elo B virus-associated multistep hepatocarcinogenesis. Mod. Pathol., 22,
meric and subtelomeric regions usingin vivo models and/or 1121–1131.
18. Plentz , R.R. et al. ( 2007 ) Telomere shortening and inactivation of cell 34 . Heagerty , P.J. et al. ( 2000 ) Time-dependent ROC curves for censored surcycle checkpoints characterize human hepatocarcinogenesis. Hepat-ol vival data and a diagnostic marker . Biometrics , 56 , 337 - 344 . ogy, 45 , 968 - 976 . 35 . Wen , C.P. et al. ( 2012 ) Hepatocellular carcinoma risk prediction model
19. Wiemann , S.U. et al. ( 2002 ) Hepatocyte telomere shortening and sene-s for the general population: the predictive power of transaminases. J. cence are general markers of human liver cirrhosis . FASEB J ., 16 , 935 - Natl . Cancer Inst. , 104 , 1599 - 1611 . 942 . 36. Chen , C.J. et al.; REVEAL-HBV Study Group . ( 2006 ) Risk of hepatocellular
20. Liu , J. et al. ( 2011 ) Longer leukocyte telomere length predicts increased carcinoma across a biological gradient of serum hepatitis B virus DNA risk of hepatitis B virus-related hepatocellular carcinoma: a case--con level . JAMA , 295 , 65 - 73 . trol analysis . Cancer , 117 , 4247 - 4256 . 37 . Schwarzenbach , H.et al. ( 2011 ) Cell-free nucleic acids as biomarkers in
21. Wan , S. et al. ( 2012 ) Telomere length in circulating serum DNA as a cancer patients . Nat. Rev. Cancer , 11 , 426 - 437 . novel non-invasive biomarker for cirrhosis: a nested case-control 38 . Blasco , M.A. ( 2005 ) Telomeres and human disease: ageing, cancer and analysis . Liver Int ., 32 , 1233 - 1241 . beyond. Nat. Rev. Genet ., 6 , 611 - 622 .
22. Fu , X. et al. ( 2012 ) Relative telomere length: a novel non-invasive 39 . Hou , L. et al. ( 2012 ) Surrogate tissue telomere length and cancer risk: biomarker for the risk of non-cirrhotic hepatocellular carcinoma in shorter or longer? Cancer Lett ., 319 , 130 - 135 . patients with chronic hepatitis B infection . Eur. J. Cancer , 48 , 1014 - 40 . Pooley, K. A .et al. ( 2010 ) Telomere length in prospective and retrospe-c 1022. tive cancer case-control studies . Cancer Res. , 70 , 3170 - 3176 .
23. Wentzensen , I.M. et al. ( 2011 ) The association of telomere length and 41 . Lee , I.M. et al. ( 2010 ) Mean leukocyte telomere length and risk of in-ci cancer: a meta-analysis . Cancer Epidemiol . Biomarkers Prev., 20 , 1238 - dent colorectal carcinoma in women: a prospective, nested case-co-n 1250 . trol study. Clin. Chem . Lab. Med., 48 , 259 - 262 .
24. Ma , H. et al. ( 2011 ) Shortened telomere length is associated with 42 . Zee , R.Y. et al. ( 2009 ) Mean telomere length and risk of incident co-lo increased risk of cancer: a meta-analysis . PLoS One , 6 , e20466. rectal carcinoma: a prospective, nested case-control approach . Cancer
25. Aviv , A. et al. ( 2009 ) Leukocyte telomere dynamics: longitudinal findings Epidemiol . Biomarkers Prev., 18 , 2280 - 2282 . among young adults in the Bogalusa Heart Study . Am. J. Epidemiol ., 43 . Farazi, P.A. et al. ( 2006 ) Hepatocellular carcinoma pathogenesis: from 169 , 323 - 329 . genes to environment. Nat. Rev. Cancer , 6 , 674 - 687 .
26. Chen , W. et al. ( 2011 ) Longitudinal versus cross-sectional evaluations of 44 . Huang , G.T. et al. ( 1998 ) Telomerase activity and telomere length in leukocyte telomere length dynamics: age-dependent telomere sho-rt human hepatocellular carcinoma . Eur. J. Cancer , 34 , 1946 - 1949 . ening is the rule . J. Gerontol. A. Biol. Sci. Med . Sci., 66 , 312 - 319 . 45 . Kojima , H. et al. ( 1997 ) Telomerase activity and telomere length in
27. Svenson , U. et al. ( 2011 ) Blood cell telomere length is a dynamic feature. hepatocellular carcinoma and chronic liver disease . Gastroenterology, PLoS One , 6 , e21485 . 112 , 493 - 500 .
28. Nordfjäll , K. et al. ( 2009 ) The individual blood cell telomere attrition rate 46 . Satyanarayana , Ae.t al. ( 2004 ) Telomeres and telomerase: a dual role in is telomere length dependent . PLoS Genet ., 5 , e1000375. hepatocarcinogenesis. Hepatology , 40 , 276 - 283 .
29. Biegler , K. A .et al. ( 2012 ) Longitudinal change in telomere length and 47 . Konishi , I. et al. ( 2009 ) Diabetes pattern on the 75 g oral glucose to-ler the chronic stress response in a randomized pilot biobehavioral cl-ini ance test is a risk factor for hepatocellular carcinoma in patients with cal study: implications for cancer prevention . Cancer Prev. Res. (Phila) ., hepatitis C virus. Liver Int ., 29 , 1194 - 1201 . 5, 1173 - 1182 . 48 . Durazo , F.A. et al. ( 2008 ) Des-gamma-carboxyprothrombin, alpha-feto-
30. Kimura , M. et al. ( 2011 ) Measurement of telomere DNA content by dot protein and AFP-L3 in patients with chronic hepatitis, cirrhosis and blot analysis . Nucleic Acids Res ., 39 , e84. hepatocellular carcinoma. J. Gastroenterol. Hepatol. , 23 , 1541 - 1548 .
31. Sanchez-Espiridion , B. et al. ( 2014 ) Telomere length in peripheral blood 49 . Welzel , T.M. et al. ( 2013 ) Population-attributable fractions of risk -fac leukocytes and lung cancer risk: a large case-control study in Ca-uca tors for hepatocellular carcinoma in the United States . Am. J. Gas-tro sians . Cancer Res. , 74 , 2476 - 2486 . enterol., 108 , 1314 - 1321 .
32. Royston , P. ( 2000 ) A strategy for modelling the effect of a continuous 50 . Fleischhacker , M. et al. ( 2007 ) Circulating nucleic acids (CNAs) and ca-n covariate in medicine and epidemiology . Stat. Med ., 19 , 1831 - 1847 . cer-a survey . Biochim. Biophys. Acta , 1775 , 181 - 232 .
33. Hann , H.W. et al. ( 2012 ) Comprehensive analysis of common serum 51 . Conomos , D. et al. ( 2012 ) Variant repeats are interspersed throughout liver enzymes as prospective predictors of hepatocellular carcinoma the telomeres and recruit nuclear receptors in ALT cells . J. Cell Biol ., in HBV patients. PLoS One , 7 , e47687 . 199 , 893 - 906 .