HPV16-E6 mRNA is superior to Cytokeratin 19 mRNA as a molecular marker
for the detection of disseminated tumour cells in sentinel lymph nodes
of patients with cervical cancer by quantitative reverse-transcription PCR
Norman H€ afner1, Mieczyslaw Gajda2, Christopher Altgassen1,3, Hermann Hertel1,4, Christiane Greinke1, Peter Hillemanns4,
Achim Schneider5and Matthias D€ urst1*
1Frauenklinik der Friedrich-Schiller-Universit€ at Jena, Jena, Germany
2Institut f€ ur Pathologie, Klinikum der Friedrich-Schiller-Universit€ at Jena, Jena, Germany
3Klinik f€ ur Frauenheilkunde und Geburtshilfe, Universit€ atsklinikum Schleswig-Holstein, Campus L€ ubeck, L€ ubeck, Germany
4Frauenklinik der Medizinischen Hochschule Hannover, Hannover, Germany
5Frauenklinik mit Hochschulambulanz der Charit? e, Campus Benjamin Franklin und Campus Mitte, Berlin, Germany
About 10–15% of patients with cervical cancer suffer from recur-
rence despite histologically negative lymph nodes (pN0). Occult
micrometastases or small tumour cell clusters may contribute to
disease outcome. The aim of this study was to compare at the RNA
level 2 known tumour-associated genes, HPV16-E6 and cytokeratin
19 (CK19), as molecular markers for the detection of disseminated
tumour cells. Real-time reverse transcription PCR technology was
used to quantify gene expression in histologically positive and nega-
tive sentinel lymph nodes (SLN) from 70 patients with cervical can-
cer. Lymph nodes from noncancer patients were used as controls.
Calculated copy numbers were normalised to the geometric average
of the most stable housekeeping genes. We observed a good correla-
tion (R 5 0.915) between the expression of both markers in SLN
with histologically confirmed metastases. However, marker gene
expression differed considerably in histologically negative nodes:
CK19 transcripts were detected in 90 of 112 SLN (80.4%), whereas
only 38 nodes (33.9%) were positive for HPV16 E6 mRNA. In par-
ticular, 62 of 74 SLN, which were negative by histology, and HPV16
E6 mRNA expressed CK19 mRNA. Moreover, 8 of 10 lymph nodes
from noncancer patients expressed CK19 mRNA. Systematic errors
due to RNA degradation or incomplete cDNA could be ruled out. It
is concluded that HPV16 E6 mRNA is more specific and more sensi-
tive for the detection of tumour cells in SLN than CK19 mRNA. The
specificity of CK19 is limited because of low level expression in unin-
volved pelvic lymph nodes.
' 2007 Wiley-Liss, Inc.
Key words: cervical cancer; disseminated tumour cells; molecular
markers; HPV16-E16; cytokeratin 19; quantitative RT-PCR
Cervical cancer is the second most common cancer among
women world wide, accounting for 493,000 new cases in 2002.1
The incidence rates vary considerably between different geo-
graphic regions, which reflect in part the lack of screening pro-
grams in developing countries. Despite the advancement in early
diagnosis, improved surgical techniques and adjuvant therapies,
approximately 40% of cervical carcinoma patients in countries
with a high standard of health care die from the disease.1The ma-
jority of these deaths is due to metastases.2
Different parameters such as lymph node metastases, tumour
grade, lymph-vascular space involvement (LVSI) and surgical
margins are prognostic factors for cervical cancer, with lymph
node status as the most important one.3,4Patients with histologi-
cally uninvolved lymph nodes (pN0) have a better prognosis than
women with histologically confirmed lymph node metastasis
(pN1) showing 5 year survival rates of 88–93% and 40–63%,
respectively.5–7Nevertheless 15% of FIGO stage IB patients with
pathologically staged negative lymph nodes (pN0) suffer from re-
currence.8The significance of micrometastases as a predictive fac-
tor for recurrent disease is the focus of several recent and ongoing
studies. Extensive serial sectioning of lymph nodes that were
judged to be free of disease by routine diagnostic evaluation
resulted in the detection of occult but prognostic significant micro-
metastases (>0.2 mm but ?2 mm) in 9% of breast cancer patients.9
Similar studies confirmed the clinical significance of micrometasta-
ses in histologically uninvolved lymph nodes not only for breast
cancer10,11but also for melanoma patients.12Besides micrometasta-
ses, small clusters of disseminated tumour cells could also have the
potential to develop into metastases. It has been estimated that rou-
tine histopathologic examination has only a 1% chance of identify-
ing a minute tumour cell cluster less than 3 cells’ diameters.13
Moreover, the detection of micrometastases and clusters of tumour
cells is further complicated by their intranodal nonrandom distribu-
tion.14Clearly, exhaustive serial sectioning combined with hema-
toxylin–eosin staining or immunohistochemistry (IHC) for the
detection of micrometastases and tumour cell clusters is time-con-
suming and expensive. The number of lymph nodes required for
analysis may therefore be reduced by examining the first draining
lymph nodes in the vicinity of the tumour. The sentinel concept was
introduced by Morton and colleagues, which allows to identify
patients who are likely to benefit from subsequent radical lymphad-
enectomy.15Importantly Giuliano and colleagues showed that senti-
nel lymphadenectomy with multiple sectioning and immunohisto-
chemical staining increases the accuracy of axillary staging in
patients with breast cancer. Immunohistochemical examination of
the sentinel node (SLN) identified significantly more patients with
metastases, especially micrometastases, than standard axillary
lymphadenectomy (42% vs. 29.1%).16For patients with cervical
cancer several studies have assessed the detection of micrometasta-
ses or disseminated tumour cells. However, the results diverge con-
siderably and the clinical significance is still unclear.17–24The aim
of this study was to evaluate the sensitivity and specificity of molec-
ular markers for the detection of tumour cells in SLN at the RNA-
rather than at the protein-level. PCR-based assays are extremely
sensitive and are less vulnerable to intranodal tumour cell distribu-
tion when compared to IHC.14,25,26It is of course critical to have a
representative amount of the SLN available for analysis which out-
weighs that of multiple sectioning. Ideally, the entire lymph node or
half of the lymph node cut along its longitudinal axis should be
used for PCR-based assays. However, for most explorative studies
such larges amounts of tissue are not available. We have used a sec-
tion of 1.0–2.0 mm thickness of each lymph node for RNA analy-
ses, which corresponds to about 200–400 serial sections.
In cervical cancer both viral- (high-risk HPV type) and tissue-
specific cellular genes are feasible markers for tumour cells. Con-
Grant sponsor: Deutsche Krebshilfe; Grant number: 70-3008-Schn5.
*Correspondence to: Frauenklink der FSU Jena, Bachstrasse 18, 07743
Jena, Germany. Fax: 149-3641-934272.
Received 9 October 2006; Accepted 6 November 2006
Int. J. Cancer: 120, 1842–1846 (2007)
' 2007 Wiley-Liss, Inc.
Publication of the International Union Against Cancer
stitutive expression of the viral oncogenes E6 and E7 is a hallmark
of cervical cancer cells.27,28Likewise CK19 is frequently used as
a tissue specific marker for disseminated cells from epithelial
tumour entities. In particular CK19 mRNA expression was
detected in 44% of histologically uninvolved lymph nodes of cer-
vical cancer patients and a high median number of transcripts cor-
related with poor prognostic outcome.17In this study we per-
formed a direct comparison of HPV16 E6 and CK19 mRNA as
molecular markers for disseminated tumour cells in lymph nodes
which were negative by conventional histopathology. Lymph
nodes with cervical cancer metastasis and lymph nodes of patients
without a gynaecological malignancy served as controls.
Material and methods
All samples were collected during an ongoing prospective mul-
ticentre study that evaluates the sentinel concept in patients with
cervical cancer and the prognostic value of viral oncogene tran-
scripts in sentinel lymph nodes. The study was approved by the
ethical committee of the FSU Jena and the written informed con-
sent was obtained from all patients. Sentinel lymph nodes (SLN)
were identified using technetium labelled albumin, Patent-Blue or
a combination of both, as previously described.29Of each SLN a
tissue section of 1–2 mm thickness was taken and immediately
snap-frozen in liquid nitrogen and stored at 280?C until RNA
extraction. All SLN and other lymph nodes were examined by rou-
tine histopathology using haematoxylin–eosin staining. From
these samples we selected 70 HPV16 positive patients and ana-
lysed 140 sentinel lymph nodes in order to compare the levels of
HPV16 E6 and CK19 mRNA. Most patients (70%) had early stage
cervical cancer classified as FIGO ? IB1. The number of exam-
ined SLN ranged from 1 to 11 lymph nodes per patient but in the
majority of cases (59 patients) one to 3 SLN were analysed. Ten
lymph nodes from 4 patients without cervical cancer served as
controls. Three of these patients had histologically confirmed en-
dometriosis. The lymph nodes had to be removed during retroperi-
toneal dissection for extensive pelvic disease. One patient had no
clinical evidence of disease.
RNA extraction, reverse transcription and real-time PCR
Total RNA was extracted from 30 mg homogenised tissue using
the RNA Blood Mini Kit (Qiagen, Hilden, Germany) according to
the manufacturer’s instructions, which include DNase treatment.
RNA concentration and quality were determined by spectropho-
tometry and gel electrophoresis. One microgram total RNA was
reverse transcribed in 20 ll using oligo-dT (500 nM) or random
primers (200 ng), dNTP (500 nM each), DTT (10 mM), first strand
buffer (Invitrogen, Karlsruhe, Germany), RNaseOUT (20 units)
and SuperScriptII reverse transcriptase (200 units) (Invitrogen,
Karlsruhe, Germany). All real-time PCR experiments were run on
a RotorGene2000 (Corbett Research, Wasserburg, Germany) or
on an ABI 7300 SDS (Applied Biosystems, Darmstadt, Germany).
Reactions were performed in 25 ll volume containing: dNTP
(240 lM each), forward and reverse primer (500 nM each),
DMSO (5%), MgCl2(1.75 mM), Tris-HCl pH8.3 (10 mM), KCl
(50 mM), gelatine (0.001%) and AmpliTaqGold (1.25 U) (Applied
Biosystems, Darmstadt, Germany). Primer specific data are listed
in Table I. Depending on the assay used either SybrGreen (0.25-
fold final concentration, Molecular Probes Invitrogen, Karlsruhe,
Germany) or a TaqMan probe (200 nM final concentration) were
added. For SybrGreen assays (HPV16 E6, GAPDH, and HPRT)
the PCR steps were as follows: Initial denaturation and hot start
activation at 95?C for 10 min followed by 40 cycles of denatura-
tion phase at 95?C for 15 sec, primer specific annealing for 20 sec
at different temperatures (see Table I) and elongation at 72?C for
40 sec. Subsequently the melting temperature of the PCR product
was determined to ensure specificity. For the detection of CK19,
we used the TaqMan assay as described by van Trappen and col-
leagues. This protocol combines annealing and elongation in
1 phase at 61?C for 1 min.17All real-time PCR results were quan-
tified using the relative standard curve approach.30We used serial
dilutions from 10 to 106copies of plasmid cloned target sequences
(pCRII-TOPO, Invitrogen, Karlsruhe, Germany) as standard
curves. Of each cDNA we determined housekeeping gene and tar-
get gene expression. A set of 8 genes tested in 20 SLN samples
were used to evaluate the gene expression stability of housekeep-
ing genes in lymph nodes31according to Vandesompele and col-
leagues.32Target gene expression measured in duplicate was nor-
malized to the geometric mean of the expression of the two most
stable housekeeping genes in lymph nodes (HPRT, GAPDH).
Performance of qRT-PCR assays
Quantitative PCR is an established and reliable method permit-
ting the sensitive and specific detection of DNA or RNA species.
We have established a SybrGreen assay for HPV16 oncogenes,
termed HPV16 E6, which detects all transcript species encoding
the oncogenes E6 and E7 or E7 only. The amplicon is located at
the 50end of the viral transcript. Because cDNA synthesis is done
with oligo-dT, our approach allows the detection of undegraded
mRNA only, which reflects the presence of intact disseminated
tumour cells. Using plasmid dilutions of cloned HPV16 DNA we
determined a dynamic range of at least 6-orders of magnitude
down to 10 HPV copies (Fig. 1a). Both parameters, PCR effi-
ciency and correlation coefficient, were only marginally affected
by template type.31Experiments in which different amounts of
RNA extracted from HPV16 positive tumour cells (SiHa) were
diluted in RNA from cells that are devoid of HPV16 E6 and CK19
mRNA (C33A) revealed a sensitivity of 1 cell (10pg) in a back-
ground of 105cells. The performance of both PCR assays is simi-
TABLE I – PRIMER DESIGN FOR QUANTITATIVE REVERSE TRANSCRIPTION PCR
F: forward, R: reverse, P: TaqMan probe, TA: annealing temperature.
1Genbank accession numbers and nucleotide (nt) position of amplicons.–2van Trappen et al.17
HPV16-E6 mRNA AND DETECTION OF TUMOUR CELLS IN SLN
lar (Fig. 1b). However, the sensitivity for the detection of SiHa
cells based on CK19 mRNA is 10 times lower when compared to
Marker gene expression in lymph nodes
All sentinel lymph nodes with histologically confirmed metasta-
ses (H1, n 5 28) were positive for HPV16 E6, whereas two of the
H1 samples were CK19 negative. The calculated median number
of transcripts in 50 ng RNA was 13443 and 80857 for HPV16 E6
and CK19, respectively (Fig. 2). Furthermore, we analysed 112
histologically negative lymph nodes (H2). In 38 of these 112
nodes (33.9%) we detected HPV16 E6 mRNA, whereas 90 nodes
(80.4%) were CK19 positive. To relate viral and CK19 transcripts
we divided the histologically negative nodes into a HPV16 E6
positive (H2/E61) and HPV16 E6 negative (H2/E62) group
(Fig. 2). The H2/E61 group showed a much lower median num-
ber of viral transcripts than H1 nodes (6 vs. 13443 copies). Simi-
larly, the CK19 expression levels were also much lower (55 vs.
70437 copies). Expression levels vary strongly and there is consid-
erable overlap between the H2 and H1 group (Fig. 2). In contrast
to the observed consistency between HPV16 E6 and CK19 levels
in H2/E61 group, 86% lymph nodes in the H2/E62 group (n 5
74) express CK19 mRNA but no viral mRNA. Moreover, the me-
dian CK19 copy number in the H2/E62 group was even slightly
higher than that in the H2/E61 samples (103 vs. 55 copies,
respectively). Notably, 8 of 10 pelvic lymph nodes from 4 patients
without cervical cancer showed a comparable expression of CK19
with a median copy number of 62 copies. These controls were also
negative for HPV16 E6 mRNA (Fig. 2).
Validation of real time PCR results
Because slight differences in PCR efficiencies of both assays
could not explain the higher detection rate for CK19, we wanted
to exclude a bias due to RNA degradation or incomplete cDNA
synthesis. These intrinsic sources of error could lead to an under-
estimation of complex mRNA species (> 1.5 kb) that are ampli-
fied at their 50end as is the case for HPV16 E6. In contrast the
PCR primers for CK19 are located within 900 nucleotides of the
polyadenylation site. For validation we selected a subset of H2/
E61/CK191 and H2/E62/CK191 lymph nodes. Two real time
PCR assays both exhibiting the same PCR efficiencies were used
to amplify the corresponding 50and 30cDNA ends of GAPDH.33
The calculated median 50-30ratios of 0.44 and 0.40 were similar
for E61 and E62 samples, respectively. Moreover, Mann–Whit-
ney analyses (a 5 0.05) revealed no significant difference
between both groups. This indicates that the RNA quality is com-
parable and rules out an underestimation of HPV16 E6 transcripts
in H2/E62/CK191 samples (Table II and Fig. 3).
To substantiate these results we also changed the cDNA pri-
ming strategy: cDNA synthesis for 13 H2/E62/CK191 samples
and 14 control samples was done by random priming. The controls
comprise 5 H1/E61/CK191, 5 H2/E61/CK191 and 4 H2/
E62/CK192 samples. Essentially no differences in expression
levels were noted irrespective of the type of cDNA synthesis per-
formed. None of the lymph nodes had to be reclassified. Impor-
tantly, all of the 13 H2/E62/CK191 samples retained their nega-
FIGURE 1 – Performance of the CK19 and HPV16 E6 PCR assays based (a) on plasmid DNA (b) on RNA extracted from the cervical carci-
noma cell line SiHa diluted in 1 lg of total cellular RNA devoid of CK19 and HPV16 E6 transcripts (C33A). 10–30 pg of RNA correspond to 1
FIGURE 2 – Box plot analyses of HPV16 E6 and CK19 mRNA lev-
els in lymph nodes with metastases (H1), in histopathologically nega-
tive but HPV16 E6 positive lymph nodes (H2/E61), in histopatho-
logically and HPV E6 negative lymph nodes (H2/E62) and in pelvic
lymph nodes of patients with benign disease (controls).
TABLE II – DETERMINATION OF RNA QUALITY IN HPV16 E6 POSITIVE
AND NEGATIVE LYMPH NODES
Lymph node group E6 CK1950-30ratio
H2/E61/CK191 (n 5 10)
H2/E62/CK191 (n 5 14)
Median transcript levels are given for HPV16 E6 and CK19. The 50-
30ratiorefers tothe median value ofthe ratioofGAPDH cDNA ends.
H€AFNER ET AL.
tive HPV16 E6 status (Fig. 4). The observed differences in marker
gene expression in H- lymph nodes, i.e. CK19 expression only, is
not caused by an underestimation of HPV16 E6 transcripts due to
RNA degradation or incomplete cDNA synthesis.
The detection of micrometastasis and tumour cell clusters in
lymph nodes of patients with cervical cancers may prove to be of
importance for disease management in the future. One of the key
factors for success is the availability of highly sensitive and spe-
cific molecular markers. Besides the definition of reliable tumour
cell markers the detection system to be used is also decisive. One
such approach is immunohistochemical analysis (IHC) after serial
sectioning of lymph nodes. Although the morphological informa-
tion is valuable, IHC is very time-consuming and expensive. As
an alternative, quantitative PCR technology has recently attracted
much interest. In this study we have compared two markers,
HPV16 E6 and CK19, for the detection of disseminated tumour
cells at the RNA level. The HPV oncogene transcripts represent
an ideal marker because they are a prerequisite for tumour pro-
gression and are required for the maintenance of the transformed
phenotype. On the other hand there are several reports in support
of CK19 mRNA as a specific marker for metastases.
We have determined the mRNA levels of HPV16 E6 and CK19
in 140 sentinel lymph nodes from 70 HPV16 positive cervical can-
cer patients. All of the histologically positive lymph nodes (H1)
were positive for HPV16 E6, whereas the two H1 samples with
the lowest E6 copy number were CK19 negative. Nevertheless
H1 nodes showed high median copy numbers of each transcript
(13443 and 80857 copies for E6 and CK19, respectively). The cal-
culated log-transformed expression values for HPV16 E6 and
CK19 showed a good correlation (R 5 0.915). The performed log
transformation is consistent with the observed log-normal distribu-
tion of mRNA levels.34Regardless of the same histological classi-
fication we observed a broad distribution of transcript copy num-
bers. This may be due to a nonuniform intranodal distribution of
tumour cells, different levels of gene expression in the tumours
per se or highly variable numbers of tumour cells in pathologically
conspicuous lymph nodes. Since the histological diagnosis of all
lymph nodes was based on HE-stained paraffin sections an exact
correlation with the molecular data that required fresh tissue is not
possible. However, carbon dye mapping of sentinel lymph nodes
from melanoma patients revealed intranodal compartmentalisation
and a non-random tumour cell distribution.14Furthermore we
could show in previous studies that the HPV16 E6 mRNA levels
vary considerably even after normalising for the fraction of
tumour cells in the biopsy (data not shown). In histologically neg-
ative nodes (n 5 112) we observed striking differences in the
expression of both marker genes. Whereas 38 of 112 (33.9%)
nodes showed HPV16 E6 expression, we detected CK19 tran-
scripts in 90 of 112 (80.4%) nodes. In particular CK19 was
expressed in 62 of 74 HPV16 E6 negative lymph nodes. More-
over, we detected CK19 with similar median copy numbers in
lymph nodes from patients without malignant disease. Although
we cannot exclude the presence of endometrial cells in pelvic
lymph nodes from endometriosis patients, it is of note that the
highest CK19 copy number (169 copies per 50 ng RNA) was
detected in a lymph node from a patient without endometriosis.
This observation suggests that the low numbers of CK19 tran-
scripts in H2/E62 lymph nodes are not tumour cell related but
reflect background levels of gene activity. The detection of the
CK19 pseudogenes ck19a and ck19b can be ruled out as we have
used the protocol of van Trappen and colleagues.17However, high
rates of false positive control samples using tissue specific
markers, including CK19, have been reported previously.35–38
This is often due to illegitimate transcription.39It is, thus, not pos-
sible to discriminate between low levels of illegitimate expression
in normal cells from high levels of expression in a few tumour
cells. Furthermore it is known that cytokines can induce CK19
expression in haemopoietic tissue in vitro and in vivo.40This is
underscored by immunohistochemical analyses, which revealed an
expression of CK19 in endothelial cells, smooth muscle cells and
in the human thymus.41–43If we account for this background gene
activity by calculating a cutoffvalue for CK19 (cutoff5 median
expression in controls 1 3*SD 5 220 copies) 63% of H2/E62/
CK191 lymph nodes would become CK19 negative. However, at
the same time 31% of the HPV16 E6 positive lymph nodes would
become CK19 negative.
Our control experiments document that the quantitative reverse-
transcription PCR assays for HPV16 E6 and CK19 mRNA have a
similar performance. Moreover, we can demonstrate that the dif-
ferences in the detection rates of viral and CK19 markers are not
due to differences in RNA quality. We conclude that HPV16 E6 is
more specific and more sensitive for the detection of intact tumour
cells in lymph nodes than CK19. We are currently evaluating the
prognostic significance of HPV oncogene transcripts in sentinel
lymph nodes of over 400 patients whose systematic lymphadenec-
tomy was negative by conventional histology. The results of this
multicentre study are expected by 2009.
FIGURE 3 – Determination of 50-30ratios of GAPDH mRNA in
HPV16 E6 positive and HPV16 E6 negative lymph nodes.
FIGURE 4 – Validation of HPV16 E6 negative lymph nodes using
poly-dT or random priming strategies for cDNA synthesis. Left panel:
quantitative reverse transcription PCR for HPV16 E6. Right panel:
quantitative reverse transcription PCR for CK19. The control group
CK192 lymph nodes
HPV16-E6 mRNA AND DETECTION OF TUMOUR CELLS IN SLN
References Download full-text
1.Parkin DM, Bray F, Ferlay J, Pisani P. Global cancer statistics, 2002.
CA Cancer J Clin 2005;55:74–108.
Fidler IJ. The pathogenesis of cancer metastasis: the ?seed and soil?
hypothesis revisited. Nat Rev Cancer 2003;3:453–8.
Hale R, Fox H, Buckley CH. Classification of cervical carcinoma.
Yuan C, Wang P, Lai C, Tsu E, Yen M, Ng H. Recurrence and sur-
vival analyses of 1,115 cervical cancer patients treated with radical
hysterectomy. Gynecol Obstet Invest 1999;47:127–32.
Alvarez RD, Soong SJ, Kinney WK, Reid GC, Schray MF, Podratz
KC, Morley GW, Shingleton HM. Identification of prognostic factors
and risk groups in patients found to have nodal metastasis at the time
of radical hysterectomy for early-stage squamous carcinoma of the
cervix. Gynecol Oncol 1989;35:130–5.
Kinney WK, Hodge DO, Egorshin EV, Ballard DJ, Podratz KC. Sur-
gical treatment of patients with stages IB and IIA carcinoma of the
cervix and palpably positive pelvic lymph nodes. Gynecol Oncol
Cheng X, Cai S, Li Z, Tang M, Xue M, Zang R. The prognosis of
women with stage IB1-IIB node-positive cervical carcinoma after rad-
ical surgery. World J Surg Oncol 2004;2:47.
Delgado G, Bundy B, Zaino R, Sevin BU, Creasman WT, Major F.
Prospective surgical-pathological study of disease-free interval in
patients with stage IB squamous cell carcinoma of the cervix: a Gyne-
cologic Oncology Group study. Gynecol Oncol 1990;38:352–7.
IBCS-Group. Prognostic importance of occult axillary lymph node
micrometastases from breast cancers. International (Ludwig) Breast
Cancer Study Group. Lancet 1990;335:1565–8.
10. Masuda N, Tamaki Y, Sakita I, Ooka M, Ohnishi T, Kadota M, Ari-
take N, Okubo K, Monden M. Clinical significance of micrometasta-
ses in axillary lymph nodes assessed by reverse transcription-polymer-
ase chain reaction in breast cancer patients. Clin Cancer Res 2000;
11. Umekita Y, Ohi Y, Sagara Y, Yoshida H. Clinical significance of
occult micrometastases in axillary lymph nodes in ‘‘node-negative’’
breast cancer patients. Jpn J Cancer Res 2002;93:695–8.
12. Bostick PJ, Morton DL, Turner RR, Huynh KT, Wang HJ, Elashoff R,
Essner R, Hoon DS. Prognostic significance of occult metastases de-
tected by sentinel lymphadenectomy and reverse transcriptase-poly-
merase chain reaction in early-stage melanoma patients. J Clin Oncol
13. Gusterson B, Ott R. Occult axillary lymph-node micrometastases in
breast cancer [letter]. Lancet 1990;336:434–5.
14. Morton DL, Hoon DS, Cochran AJ, Turner RR, Essner R, Takeuchi H,
Wanek LA, Glass E, Foshag LJ, Hsueh EC, Bilchik AJ, Elashoff D,
et al. Lymphatic mapping and sentinel lymphadenectomy for early-
stage melanoma: therapeutic utility and implications of nodal micro-
anatomy and molecular staging for improving the accuracy of detec-
tion of nodal micrometastases. Ann Surg 2003;238:538–49; discus-
15. Morton DL, Wen DR, Wong JH, Economou JS, Cagle LA, Storm FK,
Foshag LJ, Cochran AJ. Technical details of intraoperative lymphatic
mapping for early stage melanoma. Arch Surg 1992;127:392–9.
16. Giuliano AE, Dale PS, Turner RR, Morton DL, Evans SW, Krasne
DL. Improved axillary staging of breast cancer with sentinel lymphad-
enectomy. Ann Surg 1995;222:394–9; discussion 99–401.
17. van Trappen PO, Gyselman VG, Lowe DG, Ryan A, Oram DH, Bosze
P, Weekes AR, Shepherd JH, Dorudi S, Bustin SA, Jacobs IJ. Molecu-
lar quantification and mapping of lymph-node micrometastases in cer-
vical cancer. Lancet 2001;357:15–20.
18. Yuan CC, Wang PH, Ng HT, Li YF, Huang TS, Chen CY, Tsai LC,
Shyong WY. Detecting cytokeratin 19 mRNA in the peripheral blood
cells of cervical cancer patients and its clinical-pathological correla-
tion. Gynecol Oncol 2002;85:148–53.
19. Mitsuhashi A, Tanaka N, Suzuka K, Matsui H, Seki K, Sekiya S.
Detection of epidermal growth factor receptor mRNA in peripheral
blood of cervical cancer patients. Gynecol Oncol 2003;89:480–5.
20. Czegledy J, Iosif C, Forslund O, Willen R, Hansson BG. Detection of
human papilloma virus DNA in lymph nodes extirpated at radical sur-
gery for cervical cancer is not predictive of recurrence. J Med Virol
21. Juretzka MM, Jensen KC, Longacre TA, Teng NN, Husain A. Detec-
tion of pelvic lymph node micrometastasis in stage IA2-IB2 cervical
cancer by immunohistochemical analysis. Gynecol Oncol 2004;93:
22. Lentz SE, Muderspach LI, Felix JC, Ye W, Groshen S, Amezcua CA.
Identification of micrometastases in histologically negative lymph
nodes of early-stage cervical cancer patients. Obstet Gynecol 2004;103:
23. Marchiole P, Buenerd A, Benchaib M, Nezhat K, Dargent D, Mathe-
vet P. Clinical significance of lympho vascular space involvement and
lymph node micrometastases in early-stage cervical cancer: a retro-
spective case-control surgico-pathological study. Gynecol Oncol 2005;97:
24. Chan PK, Yu MM, Cheung TH, To KF, Lo KW, Cheung JL, Tong
JH. Detection and quantitation of human papillomavirus DNA in pri-
mary tumour and lymph nodes of patients with early stage cervical
carcinoma. J Clin Virol 2005;33:201–5.
25. Blaheta HJ, Roeger S, Sotlar K, Schittek B, Breuninger H, Bueltmann
B, Garbe C. Additional reverse transcription-polymerase chain reac-
tion of peripheral slices is not superior to analysis of the central slice
in sentinel lymph nodes from melanoma patients. Br J Dermatol 2004;
26. Blaheta HJ, Schittek B, Breuninger H, Maczey E, Kroeber S, Sotlar
K, Ellwanger U, Thelen MH, Rassner G, Bultmann B, Garbe C.
Lymph node micrometastases of cutaneous melanoma: increased sen-
sitivity of molecular diagnosis in comparison to immunohistochemis-
try. Int J Cancer 1998;79:318–23.
27. Walboomers JM, Jacobs MV, Manos MM, Bosch FX, Kummer JA,
Shah KV, Snijders PJ, Peto J, Meijer CJ, Munoz N. Human papillo-
mavirus is a necessary cause of invasive cervical cancer worldwide. J
28. von Knebel Doeberitz M, Rittmuller C, zur Hausen H, Durst M. Inhi-
bition of tumorigenicity of cervical cancer cells in nude mice by HPV
E6-E7 anti-sense RNA. Int J Cancer 1992;51:831–4.
29. Malur S, Krause N, Kohler C, Schneider A. Sentinel lymph node
detection in patients with cervical cancer. Gynecol Oncol 2001;80:
30. Giulietti A, Overbergh L, Valckx D, Decallonne B, Bouillon R,
Mathieu C. An overview of real-time quantitative PCR: applications
to quantify cytokine gene expression. Methods 2001;25:386–401.
31. D€ urst M, Altgassen C, Mueller B, Greinke C, H€ afner N, Schneider A:
Evaluation of HR-HPV-E6-E7 mRNA in histologically negative senti-
nel lymph nodes of cervical carcinoma patients as a predictive marker
for recurrence (Pfaffl MW, ed.), 1st International qPCR Symposium,
March 3–6, 2004; Freising-Weihenstephan, Germany, 2004. p 60.
32. Vandesompele J, De Preter K, Pattyn F, Poppe B, Van Roy N, De
Paepe A, Speleman F. Accurate normalization of real-time quantita-
tive RT-PCR data by geometric averaging of multiple internal control
genes. Genome Biol 2002;3:RESEARCH0034.
33. Swift GH, Peyton MJ, MacDonald RJ. Assessment of RNA quality by
semi-quantitative RT-PCR of multiple regions of a long ubiquitous
mRNA. Biotechniques 2000;28:524,26,28,30–1.
34. Bengtsson M, Stahlberg A, Rorsman P, Kubista M. Gene expression
profiling in single cells from the pancreatic islets of Langerhans
reveals lognormal distribution of mRNA levels. Genome Res 2005;15:
35. Merrie AE, Yun K, Gunn J, Phillips LV, McCall JL. Analysis of
potential markers for detection of submicroscopic lymph node metas-
tases in breast cancer. Br J Cancer 1999;80:2019–24.
36. Schoenfeld A, Luqmani Y, Smith D, O’Reilly S, Shousha S, Sinnett
HD, Coombes RC. Detection of breast cancer micrometastases in axil-
lary lymph nodes by using polymerase chain reaction. Cancer Res
37. Bustin SA, Siddiqi S, Ahmed S, Hands R, Dorudi S. Quantification of
cytokeratin 20, carcinoembryonic antigen and guanylyl cyclase C
mRNA levels in lymph nodes may not predict treatment failure in col-
orectal cancer patients. Int J Cancer 2004;108:412–7.
38. Bostick PJ, Chatterjee S, Chi DD, Huynh KT, Giuliano AE, Cote R,
Hoon DS. Limitations of specific reverse-transcriptase polymerase
chain reaction markers in the detection of metastases in the lymph
nodes and blood of breast cancer patients. J Clin Oncol 1998;16:
39. Chelly J, Concordet JP, Kaplan JC, Kahn A. Illegitimate transcription:
transcription of any gene in any cell type. Proc Natl Acad Sci USA
40. Jung R, Kruger W, Hosch S, Holweg M, Kroger N, Gutensohn K,
Wagener C, Neumaier M, Zander AR. Specificity of reverse transcrip-
tase polymerase chain reaction assays designed for the detection of
circulating cancer cells is influenced by cytokines in vivo and in vitro.
Br J Cancer 1998;78:1194–8.
41. Carley WW, Szczepanski A, Gerritsen ME. Cytokeratin expression
and hyaluronic acid production in cultures of human synovial micro-
vascular endothelial cells: influence of cytokines and growth factors.
42. Gown AM, Boyd HC, Chang Y, Ferguson M, Reichler B, Tippens D.
Smooth muscle cells can express cytokeratins of ‘‘simple’’ epithelium.
Immunocytochemical and biochemical studies in vitro and in vivo.
Am J Pathol 1988;132:223–32.
43. Shezen E, Okon E, Ben-Hur H, Abramsky O. Cytokeratin expression
in human thymus: immunohistochemical mapping. Cell Tissue Res
H€AFNER ET AL.