Quantitative loss of heterozygosity analysis for urothelial carcinoma detection and prognosis.
ABSTRACT To evaluate loss of heterozygosity (LOH) using microsatellite polymorphism analysis as a diagnostic and prognostic marker at the time of transurethral resection and as a follow-up marker preceding cystoscopic evidence of recurrence compared with cytology.
A total of 127 urothelial carcinoma (UC) patients were included. Tumors were staged and graded according to the International Union Against Cancer-tumor, node, metastases system and to the 2004 World Health Organization classification. LOH urinalysis was performed using 8 markers and marker-specific LOH thresholds. Thirty control samples, obtained from healthy volunteers, were used to determine the positive cut-off for each marker.
LOH was significantly more sensitive than cytology in low-grade (64.8% vs 38.5%, P <.001) and low-stage UC (68.6% vs 45.5%, P <.001). The cumulative sensitivity of cytology and LOH reached 74.7% (P <.001) for low-grade and 80.2% (P <.001) for low-stage tumors. Both urinary LOH at TP53 and chromosome 9p markers were associated with an increased risk of recurrence (relative risk = 1.73 [1.30-2.31], P = .0002) and occurred more frequently in the initial urine samples of patients who later relapsed from primary tumors (36.4% vs 0.0%, P <.05 and 57.6% vs 15.8%, P = .0001). Among 32 relapse patients, LOH was positive alongside cystoscopy in 25 of 32 cases and tested positive before cystoscopy detected recurrence in a further 5 of 25 cases.
UC diagnosis and monitoring would greatly benefit from supplementing conventional cytology with LOH urinalysis, using a panel of 8 microsatellite markers with specific threshold levels. Given the limitations of both cystoscopy and cytology, novel molecular markers are needed for detection and follow-up of UC.
- SourceAvailable from: James Catto[Show abstract] [Hide abstract]
ABSTRACT: The assessment of risk of recurrence and progression of bladder cancer (BC) is still rather difficult. We decided to check the rates of the changes mentioned above in the group of the Polish patients after a year-long observation and next to compare them with the results calculated in the European Organisation of Research and Treatment of Cancer (EORTC) risk tables. The tested group consisted of 91 patients who underwent transurethral resection of bladder tumour (TURBT). When being diagnosed, 60 cases were in the pTa clinical stage, whereas 30 cases were in T1. The coexisting carcinoma in situ (CIS) was observed in four cases. On the basis of the scores obtained from the EORTC tables, the patients were divided into the groups of low, intermediate or high risk of disease recurrence and progression. Recurrence was noticed in 23 patients (25%), while progression was observed in 11 patients (12.1%). The rate of the observed recurrences proved to be lower than it had been predicted in all the groups, except for one of the intermediate-risk group (score 1- 4). Moreover, the rate of the progressions predicted according to the EORTC risk tables was higher in all the risk groups. It can be noticed that the rate of real recurrences is lower than expected, whereas the rate of the observed progressions is overestimated. Partly, it could be the result of using a relatively small group of patients for observation and applying a different method of treatment.Central European journal of urology. 01/2013; 66(1):14-20.
- Value in Health 01/2005; 8(6). · 2.19 Impact Factor
Basic and Translational Science
Quantitative Loss of
Heterozygosity Analysis for Urothelial
Carcinoma Detection and Prognosis
Delphine Collin-Chavagnac, Christophe Marçais, Stephane Billon, Françoise Descotes,
Eric Piaton, Myriam Decaussin, Claire Rodriguez-Lafrasse, and Alain Ruffion
To evaluate loss of heterozygosity (LOH) using microsatellite polymorphism analysis as a
diagnostic and prognostic marker at the time of transurethral resection and as a follow-up marker
preceding cystoscopic evidence of recurrence compared with cytology.
A total of 127 urothelial carcinoma (UC) patients were included. Tumors were staged and graded
according to the International Union Against Cancer–tumor, node, metastases system and to the
2004 World Health Organization classification. LOH urinalysis was performed using 8 markers
and marker-specific LOH thresholds. Thirty control samples, obtained from healthy volunteers,
were used to determine the positive cut-off for each marker.
LOH was significantly more sensitive than cytology in low-grade (64.8% vs 38.5%, P ?.001) and
low-stage UC (68.6% vs 45.5%, P ?.001). The cumulative sensitivity of cytology and LOH
reached 74.7% (P ?.001) for low-grade and 80.2% (P ?.001) for low-stage tumors. Both urinary
LOH at TP53 and chromosome 9p markers were associated with an increased risk of recurrence
(relative risk ? 1.73 [1.30-2.31], P ? .0002) and occurred more frequently in the initial urine
samples of patients who later relapsed from primary tumors (36.4% vs 0.0%, P ?.05 and 57.6%
vs 15.8%, P ? .0001). Among 32 relapse patients, LOH was positive alongside cystoscopy in 25
of 32 cases and tested positive before cystoscopy detected recurrence in a further 5 of 25 cases.
UC diagnosis and monitoring would greatly benefit from supplementing conventional cytology
with LOH urinalysis, using a panel of 8 microsatellite markers with specific threshold levels.
Given the limitations of both cystoscopy and cytology, novel molecular markers are needed for
detection and follow-up of UC.
UROLOGY 76: 515.e1–515.e7, 2010. © 2010 Elsevier Inc.
(pTa-1, according to the International Union Against
Cancer–tumor, node, metastases system,2) represents
most of them.3Diagnosis is currently made by cystoscopy
followed by biopsy or transurethral resection (TUR). Not
only is cystoscopy invasive and expensive, it can also miss
carcinoma-in-situ (CIS), which may be indistinguishable
from the surrounding noncancerous urothelium.
ladder cancer is the second commonest cancer of
the genitourinary tract,1of which 90% are urothe-
lial carcinomas (UC). Nonmuscle invasive UC
After diagnosis, centers perform urine cytology every 3
months for 24 months, or longer for invasive tumors.
Operator dependency and poor sensitivity limit the reli-
ability of cytology for low-grade tumors.4An accurate,
noninvasive marker would make an ideal tool for primary
detection and follow-up.
Research has already evaluated many urine tests, and
several tumor-specific genetic abnormalities, such as mi-
crosatellite alterations, have been identified. Microsatel-
lites are polymorphic short-tandem DNA repeats found
throughout the genome. Microsatellite analysis (MA) in
both tumor and constitutional DNA reveals loss of het-
erozygosity (LOH) at numerous loci, for example, on
chromosome 9 during the early development of papillary
carcinoma and CIS.5,6To investigate the potential of
these markers as diagnostic and prognostic indicators, we
evaluated a panel of 8 microsatellite urinary markers in a
large, controlled cohort.
We primarily aimed to design and validate a strategy
for specific, robust, quantitative LOH analysis, which we
achieved by determining allelic imbalance (AI) cut-off
values in a control group of 30 healthy volunteers. We
From the Service de Biochimie et Biologie Moléculaire Sud, Centre Hospitalier Univer-
sitaire Lyon Sud, Chemin du Grand Revoyet, Pierre Bénite, Hospices Civils de Lyon,
France; Laboratoire de Radiobiologie Cellulaire et Moléculaire, Faculté de Médecine
Lyon-Sud, Université Lyon 1, Université de Lyon, Oullins, France; Centre de Pa-
thologie Est, Hôpital Femme-Mère-Enfant, Boulevard Pinel, Bron, France; Centre
d’Anatomopathologie Sud, Centre Hospitalier Universitaire Lyon Sud, Chemin du
Grand Revoyet, Pierre Bénite, France; and Service d’Urologie, Centre Hospitalier
Universitaire Lyon Sud, Chemin du Grand Revoyet, Pierre Bénite, France
Reprint requests: Delphine Collin-Chavagnac, Pharm.D., Service de Biochimie et
Biologie Moléculaire Sud, Centre Hospitalier Universitaire Lyon Sud, 165 Chemin du
Grand Revoyet, 69495 Pierre Bénite Cedex, Hospices Civils de Lyon, France. E-mail:
Submitted: May 7, 2009, accepted (with revisions): November 14, 2009
© 2010 Elsevier Inc.
All Rights Reserved
then determined the accuracy of LOH analysis compared
with urine cytology for UC diagnosis in 127 patients. Our
second goal was to evaluate the potential of this strategy
in predicting recurrence in a cohort followed up for a
maximum of 40 months after TUR. The third was to eval-
uate recurrence detection in parallel with or ahead of cys-
toscopy, to reduce the number of cystoscopic examinations.
MATERIAL AND METHODS
To evaluate LOH on chromosomes 4q32, 5q14, 9p21, 9p22,
9p23, 9q32, 13q12, and 17p13, we recruited all patients under-
going consecutive TUR at Lyon-Sud Hospital, Lyon, France,
between January 2005 and June 2007 for this prospective,
A total of 127 patients were enrolled and informed written
consent obtained in each case: 18 female and 109 male (sex
ratio, 6.0); mean age: 74 ? 12 years (range, 47-98). Of these, 72
had been referred for primary UC, and 55 were reviewed for
recurrence after initial TUR. An additional 28 recurrences
occurred during the study.
After expert pathology analysis, resected tumors were graded
and staged: 69 tumors were classified pTa-low grade, 10 pTa-
high grade, 3 pT1-low grade, 16 pT1-high grade, 18 pT2-T3, 11
CIS, occurring alone in 1 case but otherwise with additional
tumors (pT1-high grade in 6 cases; pTa-high grade in 2 cases;
pT2-high grade in 1 case, and pTa-low grade in 1 case) (Table 1).
This study complies with the latest version of the Declaration
of Helsinki and general guidelines for good clinical practice.
Each patient underwent standard follow-up, according to indi-
vidual characteristics and European Association of Urology
guidelines.7Patients were followed up prospectively after TUR;
urine samples were taken during check-ups between January
2006 and May 2008. Recurrence was defined as the presence of
a positive cystoscopy or positive histology. Histology was per-
formed in 80.6% of cases.
Urine and blood samples were collected from 30 healthy vol-
unteers recruited from our laboratory, as real negative controls
(21 females and 9 males; mean age, 42 ? 13 years, range,
20-60). We obtained informed, written consent in each case.
Blood samples (5 mL) were collected by venous puncture into
tubes containing ethylenediaminetetraacetic acid as a source of
constitutional DNA (control DNA). Buffy-coat cells were cen-
trifuged at 800 g for 10 minutes at ?4°C.
Individual voided urine (30-50 mL) samples were collected at
the urology department shortly before surgical resection (usu-
ally, but not exclusively, morning urine). Tubes were centri-
fuged at 800 g for 10 minutes at ?4°C, and supernatants were
discarded. Urinary sediment pellets were treated to remove
erythrocytes, and rinsed with phosphate-buffered saline. After
centrifugation, pellets were stored at ?80°C. The time between
urine collection, blood sampling, and laboratory storage varied
between 1 and 4 hours. Before centrifugation of each urine
sample, leukocyte concentration was established by microscopic
Urine samples were fixed with a Carbowax solution of 20%
polyethylene glycol 1500 (Merck, Darmstadt, Germany) in
50% ethanol. After centrifugation, pellets were mixed with
Cytolyt solution and centrifuged at 600 g for 10 minutes. After
discarding the supernatant, cell pellets were mixed with PreservCyt
solution and treated by the processor. A single pathologist stud-
ied the Thinprep slides (Hologic France SARL, Villepinte,
France) under a Leica DMRB microscope (Leica Microsystem
SAS, Nanterre, France). Urothelial cells were considered
high-grade when they displayed an increased nucleus/cyto-
plasm (N/C) ratio, hyperchromatism, and markedly irregular
nuclear borders or prominent nucleoli, and low-grade when
they formed papillary fronds, had an increased N/C ratio, and
a slightly irregular nuclear shape, or showed numerous elon-
Table 1. Pathological and biological (microsatellite analysis results) data for 155 tumors
Category Primary Tumor n ? 72
Xth Tumor n ? 83
* Percentage of informative loci.
†Percentage of informative urinary sediment with an allelic imbalance (AI) at the locus.
UROLOGY 76 (2), 2010
gated cells with slight nuclear abnormalities, as described by
Layfield et al.8We categorized cytologic results as positive or
negative for high-grade urothelial tumor cells. Normal, in-
flammatory, reactive, and degenerative urothelial findings
were considered negative.
Tumor stage and histologic grade were assessed according to the
International Union Against Cancer– tumor, node, metastases
system and 2004 World Health Organization classification.2
Histopathology served as the gold standard.
DNA extraction from exfoliated cells in urine and buffy coat
was performed with the FlexiGene DNA kit (Qiagen, GmbH,
Hilden, Germany) according to manufacturer recommenda-
tions. All the urine samples taken for UC diagnosis provided
enough DNA for a reliable polymerase chain reaction (PCR)
We analyzed a set of 8 polymorphic microsatellite markers
(Table 2) chosen for their performance on control DNA, in-
formativity, and LOH frequency in UC Primer sequences, and
marker locations were taken from the genome database at
http://www.ncbi.nlm.nih.gov. The markers which mapped to
chromosomes 4 (D4S243), 5 (ACTBP2), 9 (D9S747, D9S162,
D9S171, IFNA?), 13 (D13S802), and 17 (TP53) were used to
assess LOH at loci within or near CDKN2A, Rb, and TP53
genes. One primer from each marker pair was end-labeled with
WellRed fluorophore (Sigma-Proligo, The Woodlands, TX).
PCR was performed on 200 ng DNA with Optimase Taq
polymerase (Transgenomic, Inc., Omaha, NE) and Euroblu-
eTaq (Eurobio, France) using a T3 Biometra thermocycler. A
touchdown PCR program was used for 4 markers (D9S162,
D9S171, D9S747, and IFNA).
PCR product (1/40th) was analyzed with a laser-based cap-
illary electrophoresis system, the CEQ 8000 sequencer (Beck-
man Coulter, Inc., Fullerton, CA), following the manufacturer’s
instructions. Data were processed by the sequencer’s built-in
All data were compiled in an Excel database. Test sensitivity
was determined as the frequency of correctly identifying UC to
compare 2 percentages, we used the chi-square or Fisher’s exact
tests. P values of ?0.05 were considered significant. Odd ratios
were calculated using MedCalc Software.
Cut-off Determination for AI
AI was calculated as AI% ? ([Bb/Ba]-[Ub/Ua]) ? 100/
(Bb/Ba), where Ba and Bb represent the peak height of
the 2 alleles in blood and Ua and Ub in urine.9AI for an
individual marker was calculated only for the heterozy-
gotes; homozygotes being noninformative. To maximize
LOH assay specificity and reliability, LOH at an individ-
ual marker was considered positive when the AI ex-
ceeded the threshold derived from analyzing 30 healthy
volunteers; the cut-off values were chosen to ensure LOH
negativity for all healthy volunteers. Chosen cut-off val-
ues were as follows: 20% for TP53, D13S802, ACTBP2,
IFNA, 25% for D9S171, D9S747, D4S243, and 30% for
D9S162. We considered urinary sediments as LOH pos-
itive when we detected LOH for at least 1 microsatellite
marker. Each positive result was systematically double-
checked by reanalysis.
Sensitivity of LOH Analysis in UC Detection
We studied a total of 155 urine samples obtained imme-
diately before cystoscopy, including additional samples
from patients showing recurrence during the study (22
with 1 recurrence episode and 3 patients with 2). Urinary
LOH sediment analysis alone successfully detected tumor
abnormalities in 109 of 155 sediments (sensitivity ?
Table 2. Microsatellite markers used for urine analysis: chromosomal location, primer sequence and size of polymerase
chain reaction (PCR) product
Primer Sequences (5=-3=)
PCR Product Size
D9S747 9q32 (GATA)170-210
UROLOGY 76 (2), 2010
70.3%), whereas cytology alone revealed malignant cells
(positive cytology) in only 79 of 155 patients (sensitivity ?
51.0%, P ?0.001). Combining cytology and LOH signifi-
cantly improved sensitivity (81.3%), compared with either
technique alone (Table 3).
UC detection sensitivity in urine samples depended on
tumor stage and grade (Table 3). Detection sensitivity of
LOH analysis was significantly higher than cytology for
low-grade G1-G2 (59/91, 64.8% vs 35/91%, 38.5%;
P ?.001) and noninvasive pTa-pT1 tumors (83/121,
68.6% vs 55/121, 45.5%; P ?.001) (Table 3). No signif-
icant difference was found between the 2 methods for
high-grade (G3, CIS) and invasive tumors (pT2-pT3).
Compared with each technique alone, combined anal-
ysis dramatically improved the diagnostic sensitivity
for low-grade and pTa-pT1 tumors (74.7% and 80.2%,
Overall, LOH incidence for individual markers ranged
from 12.7% to 57.9% (Table 1).
We observed LOH on chromosome 9 (D9S162, D9S747,
D9S171, IFNA) in 51.5% of pTa/pT1 and 45.5% of pT2-
pT3 tumors, with more frequent allele loss on chromosome
9q (D9S747) than chromosome 9p. TP53 was more fre-
quently altered in pT2-pT3 (27.3%), high-grade tumors
(30.8%), and CIS (35.7%) than in pTa-pT1 (16.7%) or
low-grade tumors (8.9%).
LOH Analysis and Prediction of Recurrence
We studied the relationship between urinary sediment
LOH and recurrence in nonmuscle invasive UC to de-
termine the prognostic value of LOH analysis at the time
of TUR, Overall, local recurrence or metastasis occurred
in 62 of 89 (69.6%) patients, 33 of whom experienced
their first recurrence of a primary tumor (n ? 52) (mean
follow-up, 18.7 months), and 29 relapsed from a previ-
ously recurrent UC (n ? 37) (median, 17 months). Table
4 compares the urinary sediment LOH of patients with
and without tumor recurrence, with regard to overall
tumors, primary tumors, and tumors that had previously
relapsed (Xth tumor). We considered cases as recurrence-
free after a minimum of 2 normal cystoscopies at 6
months follow-up. Overall, urinary sediment LOH oc-
curred more often in patients with subsequent recurrence
than in those without (Table 4). The risk of recurrence
was significantly greater when the urinary sediment LOH
was detected in the overall set of markers (odds ratio
[OR] ? 3.15 [1.14-7.99], P ? .0265) and on the chro-
mosome 9p markers (D9S162 or D9S171 or IFNA)
(OR ? 7.50 [2.05-27.51], P ? .0024), and was slightly
higher though not significant for TP53 LOH (OR ? 6.42
[0.75-55.12], P ? .0902. Patients who relapsed from a
primary tumor showed significantly more frequent LOH
at both TP53 (0.0% vs 36.4%, P ?.05) and 9p markers
(48.4% vs 11.1%, P ? .001, respectively), whereas tu-
mors that had previously relapsed were associated with
more frequent LOH at specifically 9p markers (Table 4).
Positive predictive values for recurrence were 76.6% for
overall positive LOH at TUR and 93.3% and 90.9% for
LOH at TP53 and chromosome 9p markers, respectively.
The relative risk of recurrence with LOH at either TP53
or 9p markers was 1.73 (1.30-2.31, P ? .0002).
LOH Analysis for UC Surveillance
During the follow-up period, we collected a total of 73
urine samples from 42 patients. Tumors recurred in 32 of
42 patients. Cytology tested positive in 6 of 32 patients
(18.7%) at the time of recurrence; 1 of 10 patients
without evidence of recurrence tested positive on cytol-
ogy. Follow-up urine samples were LOH positive in 25 of
32 patients who had relapsed (78.1%). Five of these were
positive before subsequent diagnosis at cystoscopy (aver-
age interval: 4.7 months); 2 patients had a single LOH
positive urine sample at follow-up 3 months before re-
currence and 3 had 2 consecutive positive results. Fol-
low-up urine samples were LOH-positive in 6 of 10
patients who showed no subsequent recurrence on cys-
toscopy, one of which had borderline cystoscopy findings
(mean follow-up ? 18 months).
Table 3. Frequency of cytology abnormalities and loss of heterozygosity (LOH) in urine, classified as total urothelial
carcinoma (UC) cases and subdivided according to histological grade and pathological stage
Low grade (G1-G2)
High grade (G3-CIS)
NS, not significant.
* Comparing microsatellite abnormalities versus cytology using the Chi2test.
†Comparing combined LOH and cytology versus cytology alone using the Chi2test.
‡Comparing combined LOH and cytology versus LOH alone using the Chi2test.
UROLOGY 76 (2), 2010
Here we have evaluated in a large cohort of patients a
novel strategy to improve MA sensitivity, specificity, and
reliability by employing marker-specific LOH thresholds
determined here in 30 healthy volunteers at 8 loci. Al-
though several recent studies have shown the diagnostic
and prognostic utility of urine tests based on LOH anal-
ysis, they have lacked the appropriate controls to estab-
lish serious candidates for clinical biomarkers. In these
studies, thresholds were either arbitrarily chosen or val-
idated on patients with a history of UC but with negative
histology at follow-up, thus diminishing specificity. In
contrast, thresholds determined in the present study en-
sured negative LOH in all healthy controls and thus
100% specificity in UC patients.
We observed LOH frequencies on individual markers
that agree with the average 25% alteration rate per
microsatellite locus reported by Berger et al10as a reason-
able threshold for clinical practice. Chromosome 9 was the
most frequently involved, as previously reported6,11as it is
associated with the earliest stages of urothelial oncogene-
Our panel of 8 microsatellite markers demonstrated a
higher overall sensitivity compared with urine cytology
(70.3% vs 51.0%), as previously reported.13-17Impor-
tantly, urinary LOH analysis showed a dramatically in-
creased sensitivity, specifically for the noninvasive and
low-grade tumors that cytology detects poorly. These
early tumors carry a significant risk of recurrence and
their detection poses a real challenge. Moreover, the
combination of urinary LOH and cytology significantly
increased sensitivity to over 80%.
Our results and previous studies13-17suggest that con-
ventional cytology performs better than LOH analysis at
detecting high-grade and invasive tumors. However, here
our results also suggest that the combination of both tests
significantly increases sensitivity of their detection to
Urinary LOH analysis improved the detection of CIS;
missing only 2 of the 14 cases, whereas cytology missed 5.
Although CIS represents a sure indicator of dangerously
unstable urothelium, its detection is a challenge for both
cystoscopy, being often visually indistinguishable from
the surrounding normal bladder, and urinary cytology.
CIS is associated with the greatest likelihood of progres-
sion and death if undiagnosed and untreated, making
early detection crucial. Our results demonstrate that uri-
nary LOH tests may improve CIS detection. These re-
sults should be confirmed on a larger cohort of CIS. The
poor sensitivity of cytology with regard to CIS relates
Table 4. Loss of heterozygosity (LOH) and recurrence during follow-up in primary tumors vs tumors that had previously
relapsed (Xthtumor): 89 cases of nonmuscle invasive bladder cancer
All 8 loci
9p (3 markers)
* No recurrence after at least 6 months follow-up.
†Number of positive LOH patients (Pos)/ number of informative patients (Inf).
‡Chi2test or Fisher’s exact test, where appropriate.
UROLOGY 76 (2), 2010
probably to the use of voided cytology in this study.
Barbotage urine cytology could improve sensitivity com-
pared with voided urine which contains fewer cells for
Similar LOH results have been reported by other au-
thors13-17; however, all studies assessed LOH by evaluat-
ing electrophoresis profiles visually. Such analyses are
subjective and therefore do not produce quantified and
reliable results. Here, we confirm results from earlier
studies by using automated analysis, quantified results,
and thresholds determined using a control group. We
have produced reliable results from a large prospective
cohort of nonmuscle invasive UC patients. Multiplex
PCR and poolplexing before sequencer analysis may im-
prove the semiautomated MA analysis technique for rou-
tine diagnosis and follow-up.
Identifying patients at high risk of recurrence is an
important goal since such patients require a close moni-
toring, with more frequent follow-up. We stratified our
results from nonmuscle invasive UC according to early
relapse events and demonstrated an elevated risk of tu-
mor recurrence in patients with urinary LOH at TP53
and/or one of the 3 chromosome 9p markers tested
(D9S162, D9S171, IFNA). Detecting recurrent UC by
cystoscopy and conventional cytology is hampered by the
over-representation of small pTa low-grade lesions com-
pared with primary lesions. As this could also affect LOH
detection, we evaluated those patients referred for their
initial TUR separately from those undergoing TUR for
recurrence. LOH at one of the 3 chromosome 9p markers
predicted the recurrence of both primary tumors and
tumors that had previously relapsed, whereas TP53 only
predicted recurrence of primary tumors. Knowles18re-
cently reviewed the implications of markers targeting P53
and 9p in high-grade tumors at high risk of recurrence
and progression. High-grade UC, characterized by aggres-
sive behavior and a high propensity for remote metasta-
sis, typically involves major tumor suppressor genes, such
as TP53.19,20Our results suggest that urinary LOH at 9p
and TP53 may identify patients at high risk of recurrence,
who may then benefit from intravesical therapies such as
bacillus Calmette-Guerin therapy, the most effective in-
travesical treatment.21Thus, urinary LOH analysis could
become a useful prognostic tool to complement the his-
topathological staging system.
Reiterative follow-up cystoscopy is considered neces-
sary to detect recurrent disease. Follow-up commonly
consists of cystoscopy every 3 months for the first 2 years,
every 6 months for a year, and annually thereafter.7
However, the benefit of a fixed cystoscopy schedule for all
patients, whatever their risk of progression, deserves fur-
ther scrutiny. Only a few small studies have investigated
the molecular analysis of urine samples at follow-up. Our
32 patient tumor recurrence follow-up study showed a
78.1% recurrence detection rate using quantitative MA,
whereas cytology was positive in only 18.7% of patients
at the time of recurrence. These patients displayed
mainly nonmuscle invasive UC (30 of 32), in concor-
dance with a recent study by Van Rhijn et al22in which
18 of 24 (75.0%) patients with recurrence had a positive
MA at follow-up. Three earlier studies on much smaller
groups reported higher detection sensitivity,23-25perhaps
explained by the inclusion of muscle-invasive tumors
which carry more molecular abnormalities and are thus
easier to detect by LOH analysis.23Eleven patients in our
study tested LOH-positive at follow-up despite a negative
concomitant cystoscopy. These tumor-specific genetic al-
terations are likely the result of small undetected tumors
on the bladder urothelium shedding cells into the urine
sediment. Borderline cystoscopy findings in 1 case sup-
ports this theory which was later confirmed in 5 patients
by tumor recurrence diagnosed at subsequent cystoscopy
follow-ups, either 3 or 6 months later. Interestingly, the
3 patients who displayed early evidence of LOH in uri-
nary sediment 6 months before recurrence also tested
LOH positive at follow-up, 3 months before recurrence.
Our findings, along with others, strongly suggest that
noninvasive urinary molecular markers could identify
recurrence several months before detection at cystoscopy.
One limitation of our study is that urine collection at
follow-up cystoscopy may not have provided optimal
urothelial cell content. Analyzing the second urine sam-
ple in the morning may optimize tumor detection sensi-
UC diagnosis and monitoring would greatly benefit from
supplementing conventional cytology with MA. Estab-
lishing robust genetic marker profiles and determining
LOH thresholds for individual microsatellite markers in-
creases tumor DNA detection sensitivity and specificity
for diagnosis and follow-up. Molecular marker sets may
also help stratify recurrence risks and thus guide adjuvant
1. Parkin DM, Bray F, Ferlay J, et al. Global cancer statistics, 2002.
CA Cancer J Clin. 2005;55:74-108.
2. Sauter G, Knowles MA, Hartmann A, et al. Tumours of the
genitourinary system. In: Eble JN, Sauter G, Epstein JI, et al, eds.
World-Health Organization Classification of Tumours. Pathology and
Genetics of Tumours of the Urinary System and Male Genital Organs.
Lyon, France: IARC Press; 2004:89-158.
3. Kirkali Z, Chan T, Manoharan M, et al. Bladder cancer: epidemi-
ology, staging and grading, and diagnosis [review]. Urology. 2005;
4. Boman H, Hedelin H, Holmang S. Four bladder tumor markers
have a disappointingly low sensitivity for small size and low grade
recurrence. J Urol. 2002;167:80-83.
5. Cairns P, Shaw ME, Knowles MA. Initiation of bladder cancer may
involve deletion of a tumour-suppressor gene on chromosome 9.
6. Cairns P, Shaw ME, Knowles MA. Preliminary mapping of the
deleted region of chromosome 9 in bladder cancer. Cancer Res.
7. Oosterlinck W, Lobel B, Jakse G, et al; European Association of
Urology (EAU) Working Group on Oncological Urology. Guide-
lines on bladder cancer. Eur Urol. 2002;41:105-112.
UROLOGY 76 (2), 2010
8. Layfield LJ, Elsheikh TM, Fili A, et al; Papanicolaou Society of
Cytopathology. Review of the state of the art and recommenda-
tions of the Papanicolaou Society of Cytopathology for urinary
cytology procedures and reporting: the Papanicolaou Society of
Cytopathology Practice Guidelines Task Force. Diagn Cytopathol.
9. Schneider A, Borgnat S, Lang H, et al. Evaluation of microsatellite
analysis in urine sediment for diagnosis of bladder cancer. Cancer
10. Berger AP, Parson W, Stenzl A, et al. Microsatellite alterations in
human bladder cancer: detection of tumor cells in urine sediment
and tumor tissue. Eur Urol. 2002;41:532-539.
11. Kallioniemi A, Kallioniemi OP, Citro G, et al. Identification of
gains and losses of DNA sequences in primary bladder cancer by
comparative genomic hybridization. Genes Chromosomes Cancer.
12. Stoehr R, Zietz S, Burger M, et al. Deletions of chromosomes 9 and
8p in histologically normal urothelium of patients with bladder
cancer. Eur Urol. 2005;47:58-63.
13. von Knobloch R, Hegele A, Brandt H, et al. Serum DNA and urine
DNA alterations of urinary transitional cell bladder carcinoma
detected by fluorescent microsatellite analysis. Int J Cancer. 2001;
14. Seripa D, Parrella P, Gallucci M, et al. Sensitive detection of
transitional cell carcinoma of the bladder by microsatellite analysis
of cells exfoliated in urine. Int J Cancer. 2001;95:364-369.
15. Utting M, Werner W, Dahse R, et al. Microsatellite analysis of free
tumor DNA in urine, serum, and plasma of patients: a minimally
invasive method for the detection of bladder cancer. Clin Cancer
16. Dal Canto M, Bartoletti R, Travaglini F, et al. Molecular urinary
sediment analysis in patients with transitional cell bladder carci-
noma. Anticancer Res. 2003;23:5095-5100.
17. Turyn J, Matuszewski M, Schlichtholz B. Genomic instability anal-
ysis of urine sediment versus tumor tissue in transitional cell car-
cinoma of the urinary bladder. Oncol Rep. 2006;15:259-265.
18. Knowles MA. Molecular pathogenesis of bladder cancer. Int J Clin
19. Wu XR. Urothelial tumorigenesis: a tale of divergent pathways. Nat
Rev Cancer. 2005;5:713-725.
20. Dinney CP, McConkey DJ, Millikan RE, et al. Focus on bladder
cancer. Cancer Cell. 2004;6:111-116.
21. Böhle A, Jocham D, Bock PR. Intravesical bacillus Calmette-
Guerin versus mitomycin C for superficial bladder cancer: a formal
meta-analysis of comparative studies on recurrence and toxicity.
J Urol. 2003;169:90-95.
22. Van Rhijn BW, Lurkin I, Kirkels WJ, et al. Microsatellite analy-
sis—DNA test in urine competes with cystoscopy in follow-up of
superficial bladder carcinoma: a phase II trial. Cancer. 2001;92:768-
23. Steiner G, Schoenberg MP, Linn JF, et al. Detection of bladder
cancer recurrence by microsatellite analysis of urine. Nat Med.
24. Rouprêt M, Hupertan V, Yates DR, et al. A comparison of the
performance of microsatellite and methylation urine analysis for
predicting the recurrence of urothelial cell carcinoma, and defini-
tion of a set of markers by Bayesian network analysis. BJU Int.
25. Amira N, Mourah S, Rozet F, et al. Non-invasive molecular de-
tection of bladder cancer recurrence. Int J Cancer. 2002;101:293-
UROLOGY 76 (2), 2010