molecular quantification and mapping of lymph-node

7/25/2019 Molecular Quantification and Mapping of Lymph-node

1/6

For personal use only. Not to be reproduced without permission of The Lancet.

ARTICLES

Summary

Background A proportion of patients with cancer and lymph

nodes negative on histology will develop recurrence.

Reverse-transcriptase PCR (RT-PCR) is a highly sensitive

method for detection of lymph-node micrometastases, but

accurate quantitative assessment has been difficult.

Methods We studied primary tumours and 156 lymph nodes

from 32 patients with cervical cancer (stage IA2, IB1, and

IB2) and 32 lymph nodes from nine patients with benign

disease. A fully quantitative, real-time RT-PCR assay was

used to document absolute copy numbers of the epithelial

marker cytokeratin 19. Primers and probe were designed not

to amplify either of the two cytokeratin 19 pseudogenes.

Findings All primary tumours and histologically involved

lymph nodes (six) had more than 106 copies of cytokeratin

19 mRNA per g total RNA. Expression of cytokeratin 19 (upto 11105 copies per g RNA) was detected in 66 (44%) of150 histologically uninvolved lymph nodes, and in nodes

from 16 of 32 patients with cervical cancer. 15 of these 16

patients with evidence of micrometastases had the highest

cytokeratin 19 transcription level in a first lymph-node

drainage station (three obturator, six internal, and six

external iliac node). Transcription of cytokeratin 19 was

found at a low level in just one of 32 lymph nodes obtained

from nine patients with benign disease. Median copy

number of cytokeratin 19 transcription was significantly

higher (>103

copies) in association with adverse prognosticfeatures.

Interpretation The results suggest that about 50% of early-

stage cervical cancers shed tumour cells to the pelvic lymph

nodes. The amount of cytokeratin 19 expression was related

to clinicopathological features. Further studies are required

to document the clinical implications of molecular

micrometastases.

Lancet 2001; 357: 1520

IntroductionThe presence or absence of metastatic disease in pelviclymph nodes is an important prognostic factor in cervicalcancer, and careful histology of pelvic lymph nodes iscrucial in decision-making about subsequent adjuvanttherapy. Patients with histologically normal pelvic lymphnodes and clear surgical margins around the primarycervical tumour have a good prognosis and a low risk ofrecurrent disease. However, despite favourableprognostic features, pelvic recurrence occurs in about10% of patients in this category.1 Histologicallyundetectable or dormant micrometastases in thelymphatic system probably account for disease recurrenceafter variable disease-free intervals.2

Advances in understanding of cellular biologycombined with developments in molecular technologyhave provided new methods for detection of metastaticcancer cells, which are likely to be more sensitive thanconventional histopathology. One approach has usedPCR-based techniques to analyse lymphatic tissue forspecific somatic genetic alterations (eg, point mutations)that are known to be present in the primary cancer.3 Thisapproach is not useful in cervical cancer, because nospecific gene mutations have yet been identified that arepresent in the majority of these cancers. In cervicalcancer, molecular techniques have focused on thedetection of human papillomavirus DNA or its E6/E7-transforming gene transcripts in blood and lymphnodes.4,5 However, specific tumour DNA found in

histologically normal lymph nodes may originate fromdead cell material or macrophages, and viral DNA can befound in various cell types, which limits its usefulness as amolecular marker for micrometastases.

We have pursued a different approach, directed atidentifying ectopic expression in pelvic lymph nodes ofcytokeratin 19. This gene is transcribed specifically byepithelial cells, including those of the female genital tract,but it is not normally expressed by lymphoid cells andtherefore high expression in normal lymph nodes wouldnot be expected. Importantly, cytokeratin 19 expression6

remains stable throughout neoplastic transformation.7

Reverse-transcriptase PCR (RT-PCR) is a standard andsensitive technique of RNA analysis, but quantitativeinformation has been difficult to obtain owing to

limitations of specificity and reproducibility.8,9

To addressthese problems and the possibility of low-level andillegitimate10 transcription of cytokeratin 19 in lymphnodes, we used a fully quantitative, real-time11 RT-PCRassay combined with primers and a probe designed toprevent amplification of the two cytokeratin 19pseudogenes.12

MethodsSurgical samples

This prospective cohort study was approved by the LocalResearch Ethics Committee of the East London andCity Health Authority. Lymph-node samples wereobtained after dissection of pelvic lymph nodes inpatients undergoing surgery for cervical cancer at

Molecular quantification and mapping of lymph-node

micrometastases in cervical cancer

Philippe O Van Trappen, Valerie G Gyselman, David G Lowe, Andy Ryan, David H Oram, Peter Bosze, Anthony R Weekes,

John H Shepherd, Sina Dorudi, Stephen A Bustin, Ian J Jacobs

Academic Departments of Gynaecological Oncology

(P O Van Trappen MD, A Ryan PhD, D H Oram FRCOG,

Prof J H Shepherd FRCOG, Prof I J Jacobs MD) and Histopathology

(Prof D G Lowe MD), Queen Mary and Westfield College,

St Bartholomews Hospital, London, UK; Academic Department of

Surgery, Queen Mary and Westfield College, Royal London

Hospital, London (V G Gyselman MSc, S Dorudi PhD, S A Bustin PhD);

Department of Gynaecological Oncology, Saint Stephan Hospital,

Budapest, Hungary (Prof P Bosze MD); and Department of

Gynaecology, Harold Wood Hospital, Romford, Essex, UK

(A R Weekes FRCOG)

Correspondence to: Dr Philippe O Van Trappen, Academic

Department of Gynaecological Oncology, Queen Mary and Westfield

College, St Bartholomews Hospital, London EC1A 7BE, UK

(e-mail: [email protected])

THE LANCET Vol 357 January 6, 2001 15

Copyright 2001 All Rights Reserved


2/6


St Bartholomews Hospital, Harold Wood Hospital, andthe University Hospital of Budapest, between January,1998, and January, 2000. 156 lymph nodes and samplesfrom the primary tumour were collected from 32 patientswith early-stage cervical cancer, and the location of eachlymph node was documented. A further 32 lymph nodeswere available from nine patients who had suspectedmalignant disease, found to be benign ovarian cysts onpostoperative histology. Each lymph node was labelledand bisected. One half was snap-frozen in liquid nitrogenand stored at 70C until extraction of RNA. Theremainder was fixed in formalin and embedded inparaffin for conventional histopathology.

Procedures

Histological diagnosis and staging were based on theguidelines of the International Federation of Gynecologyand Obstetrics (FIGO). Up to three 3 m sections werecut from each node, stained with haematoxylin and eosin,and carefully reviewed by one specialist histopathologist(DGL) without knowledge of the cytokeratin 19 mRNAexpression. Histological type, histological grade, andlymphovascular space involvement (LVSI) were assessedon sections prepared from the tumour, stained withhaematoxylin and eosin.

Total cellular RNA was extracted from all samples bythe acid guanidinium isothiocyanate, phenol, chloroformmethod.13 Tissue was broken down by vortexing andhomogenisation. The RNA samples were stored in 50 Ldiethyl-pyrocarbonate-treated distilled water at 70C.

RNA samples were further purified by use of an RNeasyMini Protocol (Qiagen Ltd, Crawley, UK) and quantifiedon a Genequant spectrophotometer (Pharmacia,St Albans, UK).

Cytokeratin 19 (Genbank accession number Y00503)primers and probe were designed by use of PrimerExpress software (PE-ABI; version 1.6). The primersbind to sequences in exons 2 and 3, generating anamplicon of 75 bp, with the probe binding to both exonsof the cytokeratin 19 gene. Amplification and detectionof CK19a pseudogene14 (Genbank accession numberM33101) is unlikely, because the probe and thedownstream primer contain three and two mismatches,respectively. Amplification and detection of CK19bpseudogene12 (Genbank accession number U85961) is

impossible, because both primers have severalmismatches and the region complementary to the probehas extensive deletion (figure 1). The primers forthe housekeeping gene glyceraldehyde-3-phosphatedehydrogenase (Genbank accession number G04038)bind to sequences in exons 2 and 3, generating an

amplicon of 226 bp, and the probe binds to exon 3(table).

All 5-nuclease15 assays used a one-tube, one-enzymeRT-PCR protocol (Taqman).9 To prevent carry-over ofcontaminating DNA, the reaction was carried out in thepresence of dUTP. Before reverse transcription, the RNAtemplate was heated for 2 min at 50C in the presence of001 U/L uracil N-glycosylase. Total RNA (50100 ng)was reverse transcribed in a 25 L reaction mixture at60C for 30 min. After 5 min denaturation at 92C, PCRwas carried out for 40 cycles with denaturation at 92Cfor 20 s and extension at 62C for 60 s in the presence ofan oligonucleotide probe containing a fluorescent dye(6-carboxyfluorescein) at its 5-end and a quencher(6-carboxytetramethylrhodamine) at its 3-end.11 The

RT-PCR assay for cytokeratin 19 was done four times foreach RNA sample to assess consistency of results. Aftertarget amplification, the probe anneals to the ampliconand is displaced and cleaved between the reporter andquencher dyes by the nucleolytic activity of therecombinant Thermus thermophilus DNA polymerase.The amount of product resulting in detectablefluorescence at any given cycle within the exponentialphase of PCR is proportional to the initial number oftemplate copies. The number of PCR cycles (thresholdcycle) needed to detect the amplicon is therefore a directmeasure of template concentration. To monitorcontamination, reactions were done with two no-template controls in every amplification run, oneprepared before the tubes were opened and the other at

the end of the experiment. Reactions were recorded andanalysed with the ABI 7700 Prism Sequence detectionsystem (Perkin-Elmer Applied Biosystems, Warrington,UK). Accurate quantification was achieved throughthe generation of standard curves by serial dilutionof cytokeratin 19 and glyceraldehyde-3-phosphatedehydrogenase RNA transcribed by T7 RNApolymerase.

Statistics

Statistical analyses used SPSS software (SPSS forwindows, version 6.1.3). The Mann-Whitney Utest wasused to compare the median copy number and range ofcytokeratin 19 mRNA in histologically uninvolved lymphnodes between the different tumour grades, FIGO stages,

and presence or absence of LVSI.

ResultsTranscription of glyceraldehyde-3-phosphate dehydrog-enase was confirmed in all samples studied, and nosignificant difference was found in the median number of

ARTICLES

16 THE LANCET Vol 357 January 6, 2001

mRNA Amplification Product size (bp) Primers Probe (5-6-carboxyfluorescein-3-6-carboxytetramethylrhodamine)

Cytokeratin 19 Exons 2 and 3 75 Forward: TCGACAACGCCCGTCTG CCGAACCAAGTTTGAGACGGAACAGG

Reverse: CCACGCTCATGCGCAG

GAPDH Exons 2 and 3 226 Forward: GAAGGTGAAGGTCGGAGTC CAAGCTTCCCGTTCTCAGCC

Reverse: GAAGATGGTGATGGGATTTC

Primer and probe sequences for RT-PCR amplification of cytokeratin 19 and glyceraldehyde-3-phosphate dehydrogenase (GAPDH)

mRNA

CK19

CK19a

CK19b

exon 2 exon 3

Figure 1: Amplicon generated by amplification of cytokeratin 19 mRNAArrows identify location of upstream and downstream primers. The sequence recognised by the labelled probe is shaded. Sequences of pseudogenes

CK19a and CK19b are also shown. =deletion in CK19b compared with CK19. Vertical bar indicates exons 2 and 3 splice junction.



3/6


these transcripts in lymph nodes from patients withcervical cancer and those with benign disorders (mediancopy number 10107 vs 18107 per g RNA; p=069).

We studied primary cervical tumours from 32 patients

with cervical cancer (22 squamous-cell carcinomas, twoclear-cell carcinomas, six adenocarcinomas, and twoadenosquamous carcinomas; one FIGO stage IA2, and31 FIGO stage IB1/2) who underwent primary surgery(radical hysterectomy and pelvic lymphadenectomy).The age range was 2670 years (median 41). Cytokeratin19 transcription was detected in all primary tumours witha median copy number of 81106 per g RNA (range42105 to 58108 per g RNA).

The patients with benign disease ranged in age from 45to 66 years (median 53). Cytokeratin 19 transcriptionwas detected in just one of 32 lymph nodes, at a copynumber of only 66 per g RNA. This lymph node wasfrom a patient diagnosed with a benign serouscystadenoma of the ovary.

Six of the 156 lymph nodes from the 32 patients withcervical cancer had histological evidence of metastaticinvolvement. All six showed cytokeratin 19 transcriptionwith copy numbers similar to the primary tumour(median 7106 per g RNA [range 34106 to 56107

per g RNA). Two of the four patients with histologicallyinvolved lymph nodes showed cytokeratin 19transcription in all other lymph nodes, which werehistologically uninvolved. Both patients had poorlydifferentiated stage IB2 tumours, with LVSI.

Of the 150 histologically uninvolved lymph nodes, 66(44%) showed cytokeratin 19 transcription (median49103 per g RNA) over a range up to 11105 copynumbers per g RNA. There were significantassociations between the degree of cytokeratin 19

transcription in histologically uninvolved lymph nodesand clinicopathological prognostic features (figure 2).Cytokeratin 19 transcription was higher in the lymphnodes from patients with poorly differentiated (grade 3)tumours than in those with well or moderatelydifferentiated tumours (median copy number 38103 vs0 per g RNA; p


4/6


1103 per g RNA. In each case the proportion positivefor cytokeratin 19 expression was higher in the worseprognostic category than in that with better prognosis.

Among the 16 patients with detectable cytokeratin 19mRNA expression in their lymph nodes, the highestexpression was found in the obturator nodes in threecases, in the internal iliac nodes in six, in the externaliliac nodes in six, and in the common iliac node in one.According to anatomical location, the lowest copynumbers were found in the obturator nodes (median 0per g RNA; figure 3). These copy numbers weresignificantly lower than those in the internal iliac nodes(median 7102 per g RNA, p=001) and common iliacnodes (median 37103 per g RNA, p=0005). Theproportions positive were highest among common andinternal iliac nodes.

The mean follow-up of the patients with cervicalcancer was 21 months. Three patients developedrecurrence, one with a distant metastasis. The firstpatient developed recurrence 3 months after primarysurgery and showed recurrent disease centrally and at thepelvic side wall. This patient was initially diagnosed witha histologically node-negative, stage IB1, poorlydifferentiated squamous-cell carcinoma of the cervix

with LVSI in the primary tumour. High cytokeratin 19mRNA copy numbers (>103) were found in all sixuninvolved lymph nodes from this patient. The secondpatient developed recurrent disease with a distantmetastasis in a lymph node in the neck 12 months afterprimary surgery. This patient was initially diagnosed witha stage IB2, poorly differentiated adenocarcinoma of thecervix with LVSI in the primary tumour. Two pelviclymph nodes were histologically involved, andquantitative RT-PCR analyses revealed presence ofcytokeratin-19-positive tumour cells (with >103

transcripts per g RNA) in all six histologicallyuninvolved lymph nodes analysed. The third patient, whowas initially diagnosed with a node-negative, stage IB1,moderately differentiated, squamous-cell carcinoma of

the cervix and LVSI in the primary tumour, developedrecurrent disease 18 months after primary surgery.Quantitative RT-PCR revealed the presence ofcytokeratin-19-positive tumour cells (103 transcripts perg RNA) in an internal iliac node at the time of primarydiagnosis.

DiscussionThe concept of the sentinel node as the first drainagestation in breast cancer was introduced in 1994.16

Immunohistochemistry has been used to assess thepresence of micrometastases in nodes classified asuninvolved on conventional histology, because aproportion of patients with node-negative breast cancerwill develop recurrence.17 Rates of conversion based on

immunohistochemistry vary between 8% and 41%.18,19

The use of a more sensitive technique, RT-PCR forcytokeratin 19 mRNA, has shown even higher conversionrates in patients with node-negative breast cancer.19

However, several issues need to be addressed aboutthe significance of micrometastases. Since the firstreport in 1991 by Smith and colleagues, severalmolecular markers for micrometastases in various cancersand different PCR protocols have been described, butnone has been introduced in standard clinical practice.20,21

Nevertheless, one study has shown the clinical potentialof RT-PCR and found significant differences in 5-yearsurvival between patients with stage II colorectal cancerwith and without evidence of micrometastases, by use ofRT-PCR for expression of carcinoembryonic antigen

mRNA.22 For most cancers, however, there are nomethods yet to detect which patients withmicrometastases will develop recurrence and how thiswill affect survival, because not all micrometastases willdevelop into distant tumours. Benson and Querci dellaRovere raised the important issue of potentialoverstaging.19 Because of the increasing number ofreports on detection of micrometastases withconventional RT-PCR, new questions have arisen. Is thetotal burden of micrometastases important? How can weassess their behaviour to predict potential further growth?And do we need to remove all lymph nodes if the sentinelnode is negative? The aim of our study was to quantifythe amount of cytokeratin 19 mRNA expression with anovel molecular technique, real-time quantitative RT-PCR. We also undertook lymphatic mapping to assessthe pattern of micrometastatic spread.

We documented high transcription of cytokeratin 19 inall 32 primary cervical tumours studied. We were able toquantify accurately cytokeratin 19 transcription in lymphnodes with putative occult micrometastatic disease. Toassess the specificity of cytokeratin 19 mRNA as amolecular marker for micrometastases in lymph nodes,we analysed 32 nodes from patients with benign epithelial

disease. Only one showed transcription of cytokeratin 19and only at a very low level. This transcription could bedue to expression of cytokeratin 19 by myoepithelial cellssurrounding blood vessels present in lymph nodes, orillegitimate transcription of the gene by lymphoid cells.10

The frequency of cytokeratin 19 positivity in lymphnodes from patients without cancer is lower in our studythan reported by others.8,23 A possible explanation is thatthe recently identified second pseudogene12 was amplifiedin earlier conventional (nested) RT-PCR assays forcytokeratin 19, giving a higher false-positive rate. Wespecifically designed primers and probe not to amplifyeither of the two known pseudogenes.

50% of the patients with cervical cancer in our studyshowed cytokeratin 19 transcription over a broad range in

their lymph nodes, suggesting occult micrometastases.This proportion broadly accords with that in a study ofcolorectal cancer,24 but it is slightly higher than thatreported in some studies of breast cancer.8,23 Theinconsistency in the proportion of micrometastasesdetected in lymph nodes of cancer patients by RT-PCR-based assays has several possible explanations. First,different methods have been used for sample processingand RNA extraction. Second, RT-PCR protocols are notstandardised, and different primers for the samemolecular marker or altogether different markers havebeen used for the same cancer. Third, the number oflymph nodes analysed per patient has variedsubstantially, from three to 15. Our high proportion ofputative occult micrometastases (44%) in normal lymph

nodes from patients with cervical cancer may be relatedto the high sensitivity of real-time quantitative RT-PCR,which enables us to detect low transcription ofcytokeratin 19. The histologically involved lymph nodesin our study showed similar degrees of transcription ofcytokeratin 19 to the primary cervical tumours. Thedegree of transcription was also similar to that detected inhistologically involved lymph nodes from othergynaecological cancers (data not shown). All additionaluninvolved lymph nodes from one of the patients withnode-positive cervical cancer showed more than 103

copies of cytokeratin 19 mRNA per g total RNA. Onenode-negative patient had more than 103 copies per gRNA in all lymph nodes. Both patients developedrecurrent disease.

ARTICLES




5/6


We found significant differences in transcriptionaccording to differentiation, stage, and the absence orpresence of LVSI in the primary tumour. These threeprognostic factors are associated with clinical outcome incervical cancer.1,25 Whether the apparent threshold valueof 103 copies of cytokeratin 19 transcripts per g RNA inhistologically uninvolved lymph nodes will also indicate apoor prognosis is not yet known.

The potential first lymph-node drainage stations forcervical cancer are the obturator, parametrial, andinternal and external iliac nodes. Dargent and colleaguesused a dye solution to identify the sentinel node withlaparoscopy in 23 patients with early-stage cervicalcancer.26 They were able to identify one or two sentinelnodes per patient, mostly in the internal iliac area. Withhaematoxylin and eosin staining, they found between oneand three nodes with evidence of metastatic involvementon histology and in all these cases the sentinel node waspositive. We found no consistent anatomical location forthe lymph node with the highest cytokeratin 19transcription, although 15 of 16 patients with evidence ofmicrometastases had the highest transcription in a firstlymph-node drainage station. The finding of the highestcytokeratin 19 copy numbers in the lymph nodes most

proximal to the primary cancer supports the concept thatcervical cancers (like several other cancers) metastasisevia a sentinel lymph-node. Real-time quantitative RT-PCR might be valuable in detection of micrometastasesin sentinel nodes found after the application of a dyesolution or radioisotope. Van der Velde-Zimmermannand colleagues reported a possible strategy for analysingsentinel nodes in melanoma patients by use of RT-PCRas first diagnostic tool.27 This highly sensitive and quicktechnique can be used effectively to preselect nodes forfurther immunohistochemistry of serial sections, and itseems both rapid and cost-effective.

Slade and colleagues proposed the use of competitiveRT-PCR to improve the specificity of the conventionalcytokeratin 19 RT-PCR assay.28 Although this technique

aims to quantify cytokeratin 19 transcripts by means of acompetitor template and seems to be more specific thannested PCR, such an assay is too complicated for routineclinical use. The method we used has several advantagesas a molecular diagnostic tool compared withconventional RT-PCR. It is at least as sensitive asconventional nested RT-PCR but far more specific, itrequires no post-PCR manipulation such as gelelectrophoresis or Southern blotting, and it lends itself toeasy automation.11 Absolute quantification of mRNAcopy numbers is easily achieved by use of standard curvesfor each molecular marker used.

We have shown that 50% of patients with early-stagecervical cancer shed tumour cells in the lymphaticsystem, and that the amount is significantly associated

with poor clinicopathological prognostic features.Evidence from necropsy data and animal models suggeststhat most cancer patients have dormant micrometastases.However, only a very small fraction of thesemicrometastases continue to grow to form tumours.2,29,30

Quantification, characterisation, and assessment of thepotential of distant tumour growth of micrometastases istherefore essential in further studies on micrometastases.

The results of this study provide a proof of principle formolecular studies of micrometastases. Ultimately, a panelof RT-PCR markers including other tumour productssuch as angiogenic factors may provide optimum clinicalinformation. Such a panel may enable clinicians toidentify the small group of patients who are destined todevelop recurrent disease.

ContributorsPhilippe Van Trappen was responsible for the conception, design,

arrangement of ethical approval, and collection of surgical specimens,

conduct of molecular experiments, analyses, interpretation of the data,

and drafting of the paper. Valerie Gyselman constructed the standard

curves and was involved in technical support during the molecular

experiments. David Lowe was responsible for the histopathological

analyses. Andy Ryan was involved in practical arrangements for the

collection of tumour samples and RNA extractions. David Oram,

Peter Bosze, Anthony Weekes, and John Shepherd were involved in the

organisation of sample collections and follow-up of patients. Sina Dorudi

provided the quantitative RT-PCR facilities. Stephen Bustin designed the

primers and probes and contributed to the design of the methods.Ian Jacobs was involved in the design of the study, interpretation of the

data, and drafting of the paper.

AcknowledgmentsWe thank Janice Thomas for her assistance in the statistical analyses and

Pal Siklos for providing samples. The ABI 7700 Prism Sequence

detection system (PE-ABI) was donated by the London Immunotherapy

Cancer Trust. POVT was supported by the Luxembourg Cancer

Foundation and the Joint Research Board of the Special Trustees of St

Bartholomews Hospital, London.

References

1 Inoue T, Morita K. The prognostic significance of number of positive

nodes in cervical carcinoma stages IB, IIA and IIB. Cancer1990; 65:

192327.

2 Holmgren L, OReilly M, Folkman J. Dormancy of micrometastases:

balanced proliferation and apoptosis in the presence of angiogenesis

suppression.Nat Med 1995; 1: 14953.

3 Hayashi N, Ito I, Yanagisawa A, et al. Genetic diagnosis of lymph-

node metastasis in colorectal cancer. Lancet 1995; 345: 125759.

4 Czegledy J, Iosif C, Forslund O, Willen R, Hansson BG. Detection of

human papilloma virus DNA in lymph nodes extirpated at radical

surgery for cervical cancer is not predictive of recurrence. J Med Virol

1998; 54: 18385.

5 Pao CC, Hor JJ, Yang FP, Lin CY, Tseng CJ. Detection of human

papillomavirus mRNA and cervical cancer cells in peripheral blood of

cervical cancer patients with metastasis.J Clin Oncol 1997;

15: 100812.

6 Stasiak PC, Purkis PE, Leigh IM, Lane EB. Keratin 19: predicted

amino acid sequence and broad tissue distribution suggest it evolved

from keratinocyte keratins.J Invest Dermatol 1989; 92: 70716.

7 Moll R, Levy R, Czernobilsky B, Hohlweg-Majert P,

Dallenbach-Hellweg G, Franke WW. Cytokeratins of normal

epithelia and some neoplasms of the female genital tract. Lab Invest1983; 49: 599610.

8 Schoenfeld A, Luqmani Y, Smith D, et al. Detection of breast cancer

micrometastasis in axillary lymph nodes by using polymerase chain

reaction. Cancer Res 1994; 54: 298690.

9 Bustin SA, Gyselman VG, Williams NS, Dorudi S. Detection of

cytokeratins 19/20 and guanylyl cyclase C in peripheral blood of

colorectal cancer patients. Br J Cancer 1999; 79: 181320.

10 Chelly J, Concordet JP, Kaplan JC, Kahn A. Illegitimate

transcription: transcription of any gene in any cell type. Proc Natl

Acad Sci USA 1989; 86: 261721.

11 Heid CA, Stevens J, Livak KJ, Williams PM. Real time quantitative

PCR. Genome Res 1996; 6: 98694.

12 Ruud P, Fodstad , ,Hovig E. Identification of a novel cytokeratin 19

pseudogene that may interfere with reverse transcriptase-polymerase

chain reaction assays used to detect micrometastatic tumor cells.

Int J Cancer 1999; 80: 11925.

13 Chomczynski P, Saachi N. Single-step method of RNA isolation by

acid guanidinium thiocyanate-phenol-chloroform extraction.Anal Biochem 1987; 162: 15659.

14 Savtchenko ES, Schiff TA, Jiang CK, Freedberg IM, Blumenberg M.

Embryonic expression of the human 40-kD keratin: evidence from a

processed pseudogene sequence.Am J Hum Genet 1988; 43: 63037.

15 Holland PM, Abramson RD, Watson R, Gelfand DH. Detection of

specific polymerase chain reaction product by utilizing the 5 -3exonuclease activity of Thermus aquaticus DNA polymerase. Proc Natl

Acad Sci USA 1991; 88: 727680.

16 Giuliano AE, Kirgan DM, Guenther JM, Morton DL. Lymphatic

mapping and sentinel lymphadenectomy for breast cancer.Ann Surg

1994; 220: 391401.

17 Cote RJ, Peterson HF, Chaiwun B, et al. Role of

immunohistochemical detection of lymph-node metastases in

management of breast cancer. Lancet 1999; 354: 896900.

18 McIntosh SA, Going JJ, Soukop M, Purushotham AD, Cooke TG.

Therapeutic implications of the sentinel lymph node in breast cancer.

Lancet 1999; 354: 570.

ARTICLES

THE LANCET Vol 357 January 6, 2001 19



6/6


19 Benson JR, Querci della Rovere G. Sentinel lymph node in breast

cancer. Lancet 1999; 354: 1998.

20 Smith B, Selby P, Southgate J, Pittman K, Bradley C, Blair GE.

Detection of melanoma cells in peripheral blood by means of reverse

transcriptase and polymerase chain reaction. Lancet 1991; 338:

122729.

21 Raj GV, Moreno JG, Gomella LG. Utilization of polymerase chain

reaction technology in the detection of solid tumors. Cancers 1998;

82: 141942.

22 Liefers G-J, Cleton-Jansen A-M, van de Velde CJ, et al.

Micrometastases and survival in stage II colorectal cancer. N Engl J

Med 1998; 339: 22328.23 Schoenfeld A, Luqmani Y, Sinnett HD, Shousha S, Coombes RC.

Keratin 19 mRNA measurement to detect micrometastases in lymph

nodes in breast cancer patients. Br J Cancer 1996;

74: 163942.

24 Futamura M, Takagi Y, Koumara H, et al. Spread of colorectal

cancer micrometastases in regional lymph nodes by reverse

transcriptase-polymerase chain reactions for carcionembryonic

antigen and cytokeratin 20.J Surg Oncol 1998; 68: 3440.

25 Alvarez RD, Soong SJ, Kinney WK, et al. 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: 13035.

26 Dargent D, Martin X, Roy M, Mathevet P. Identification of a

sentinel node with laparoscopy in cervical cancer. Proc SGO Meeting

2000; abstract 44: 128.

27 van der Velde-Zimmermann D, Schipper ME, de Weger RA,

Hennipman A, Borel Rinkes IH. Sentinel node biopsies in melanoma

patients: a protocol for accurate, efficient, and cost-effective analysis

by preselection for immunohistochemistry on the basis of Tyr-PCR.

Ann Surg Oncol 2000; 7: 5154.

28 Slade MJ, Smith BM, Sinnett HD, Cross NC, Coombes RC.Quantitative polymerase chain reaction for the detection of

micrometastases in patients with breast cancer. J Clin Oncol 1999; 17:

87079.

29 Luzzi KJ, MacDonald IC, Schmidt EE, et al. Multistep nature of

metastatic inefficiency: dormancy of solitary cells after successful

extravasation and limited survival of early micrometastases.

Am J Pathol 1998; 153: 86573.

30 Cifuentes N, Pickren JW. Metastases from carcinoma of mammary

gland: an autopsy study.J Surg Oncol 1979; 11: 193205.

ARTICLES


C i ht 2001 All Ri ht R d

molecular quantification and mapping of lymph-node

Documents